Surface Tuning of Wood via Covalent Modification of Its Lignocellulosic Biopolymers with Substituted Benzoates-A Study on Reactivity, Efficiency, and Durability.

Chemical modification of wood applying benzotriazolyl-activated carboxylic acids has proven to be a versatile method for the durable functionalization of its lignocellulosic biopolymers. Through this process, the material properties of wood can be influenced and specifically optimized. To check the scope and limitations of this modification method, various benzamide derivatives with electron-withdrawing (EWG) or electron-donating (EDG) functional groups in different positions of the aromatic ring were synthesized and applied for covalent modification of Scots pine (Pinus sylvestris L.) sapwood in this study. The bonded amounts of substances (up to 2.20 mmol) were compared with the reactivity constants of the Hammett equation, revealing a significant correlation between the modification efficiency and the theoretical reactivity constants of the corresponding aromatic substitution pattern. The successful covalent attachment of the respective substituted benzamides was proven by attenuated total reflection infrared (ATR-IR) spectroscopy, while the stability of the newly formed ester bond was proven in a standardized leaching test.

Introduction

Chemical modification of wood enables a longer service lifespan of the natural material and increases the attractiveness and usability of the material.[1] Depending on the modification method, the treatment leads to an increased dimensional stability, hydrophobization and fire retardance of the biomaterial, as well as an increased resistance to pest infestation or decomposition by fungi as described in numerous reviews.[1−20] Modified wood panels are used in several fields of application like the construction of weather-prone buildings, facades, and terraces.[21] A particularly promising area of application is the modification of soft and fast-growing types of wood to optimize their properties so that they can be used as a substitute for harder and slow-growing tropical wood to protect the rainforest.[22] Therefore, the chemical modification of wood has attracted more and more interest in recent years, which is evident from a distinct increase in fundamental research and development leading to a number of industrially used processes.[1,2,13,23−29] Especially, the well-known acetylation of wood (ACCOYA process),[1,2,27,30−35] implementation with the urea derivative DMDHEU (BELMADUR process),[1,2,36−41] and impregnation with furfuryl alcohol (KEBONY process)[2,10,12,27,42,43] have been studied and optimized extensively. From a chemical point of view, these methods are fairly simple and can be easily implemented in terms of process technology: The wood modifications are carried out in vacuum tanks with subsequent drying. In addition to the technically applied acetylation, Yuan et al.[44] and Evans et al.[45] described a method for the benzoylation of wood applying benzoyl chloride. This procedure resulted in enhanced thermal stability as well as weathering and photostability of the treated meal and veneer samples. However, the chemical scope of the described processes is quite limited and unspecific so that these methods only allow fundamental protection of wood against certain harmful phenomena. A more specific optimization of the chemical wood modification is of great importance to enlarge the area of application of natural plant materials to replace nonrenewable ones.

Kaufmann and co-workers[46] introduced a mild and versatile method for the effective benzoylation of wood. The esterification of wood applying benzotriazolyl-activated carboxylic acids allows a highly functional modification of the biopolymers in contrast to the established but restricted methods described above. It is now possible to specifically influence wood hydrophobicity using halo, silyl, or alkyl benzamides,[47−49] and increase the resistance against fungi.[50] It is also possible to gain control against biological attack applying adapted insecticides[51] or adding flame retardant properties by binding boron or phosphorus derivatives.[52] In addition, it has been proven that benzotriazolyl-activated carboxylic acids are not only able to modify the surface but also to penetrate the wood material below the surface.[53,54] This versatile method for the functionalization of lignocellulosic biopolymers can be applied not only for the modification of wood but also for the adjustment of natural fibers. To further investigate this benzoylation method, extensive reactivity studies were carried out applying a variety of differently substituted benzamides at various temperatures. The ambition of this work was to gain an understanding of the reaction behavior of the specially designed derivatives to be able to develop more specific wood preservatives based on activated carboxylic acids in the future. For this purpose, several substituted benzoic acid derivatives with electron-withdrawing groups (EWGs) or electron-donating groups (EDGs) in different positions of the aromatic ring were activated with 1H-benzotriazole and subsequently used for covalent modification of pine wood chips. The bound amounts of substances were finally compared with the corresponding reactivity constants of the Hammett equation. This revealed a significant connection between the modification efficiency and the basic reactivities of comparable compounds in organic chemical esterification reactions.

The durable long-term protection is a major environmental advantage of covalent wood modification avoiding frequent reimpregnation with conventional preservatives. But chemical wood modification also has an opposing impact on the environment due to extra processing as well as life extension of the material as evaluated by Hill et al.[55] for conventional modification methods. Therefore, it is particularly important that the compounds used for chemical wood modification are stably bound to the wood biopolymers and leaching into the environment is prevented. In this study, the stability of the ester bond formed between the applied reagent and the wood biopolymers was tested as proven in a standardized leaching test according to DIN EN 84.[56]

Results and Discussion

Chemical Syntheses of the Modification Reagents

The basis for the following wood modification was the synthesis of the activated carboxylic acids with 1H-benzotriazole. This activation process was originally developed by Katritzky in which the acid group is first converted to the carboxylic acid chloride using thionyl chloride and then reacted in situ with 1H-benzotriazole to yield the activated benzotriazolate.[57]Figure provides the applied reaction conditions for the activation previously also used by the Kaufmann group.[46−51]

Figure 1

Synthesis of 1H-benzotriazole-activated benzoic acids.

This reaction was performed under mild conditions at room temperature in anhydrous dichloromethane and provided the activated amides 1–28 in very good yields (Table ). The resulting carboxamides 1–28 are characterized by high stability in combination with a high reactivity toward nucleophiles, in this case, wood hydroxy groups. In addition, the molecular structure of carboxamide 25 was verified via X-ray structure analysis (Figure ).

Table 1

1H-Benzotriazole Activation of Aromatic Carboxylic Acids—Overview of the Results

starting material	activated acid	substitution pattern	yield (%)
1a	1	4-H	84
2a	2	4-F	88
3a	3	4-Cl	94
4a	4	3-Cl	94
5a	5	2-Cl	96
6a	6	4-Br	91
7a	7	4-CN	69
8a	8	4-NO2	85
9a	9	3-NO2	86
10a	10	2-NO2	97
11a	11	4-NH2	36
12a	12	4-N(CH3)2	92
13a	13	4-Si(CH3)3	75
14a	14	4-OH	84
15a	15	4-OCH3	93
16a	16	3-Cl, 4-OH	21
17a	17	3-Cl, 4-OCH3	88
18a	18	3-NO2, 4-OH	55
19a	19	3-NO2, 4-OCH3	83
20a	20	4-SCH3	77
21a	21	4-SO2CH3	90
22a	22	4-COOCH3	83
23a	23	4-CH3	96
24a	24	4-CF3	88
25a	25	4-CH2Cl	94
26a	26	4-CH2Br	91
27a	27	4-phenyl	94
28a	28	anthracenyl-9-carboxamide	91

Figure 2

X-ray structure of activated carboxamide 25.

Wood Modification

The synthesized carboxamides were applied for wood modification using the benzoylation method developed by Kaufmann and co-workers.[46−51] Veneer chips of Scots pine sapwood (Pinus sylvestris L.) were used for the modification process. To ensure a high sample throughput, the modifications were performed in a “Synthesis 1” liquid system parallel synthesizer. A mean value of 7.0 mmol of accessible hydroxy groups per gram wood was assumed since exchange experiments with titrated water described by Sumi et al.[58] showed that an average of 6.9–8.0 mmol of hydroxy groups per gram wood was accessible for chemical reactions. The reaction was catalyzed by the bases triethylamine and 4-(N,N-dimethylamino)pyridine (DMAP). Anhydrous dimethylformamide (DMF) was used as the solvent and swelling agent since previous studies by Mantanis proved that DMF has excellent swelling properties for wood.[59,60] Before and after the modification reaction, all wood samples were subjected to Soxhlet extraction in a solvent mixture of toluene:acetone:methanol in a ratio of 4:1:1 and subsequently dried at 105 °C to remove unbound ingredients from the biomaterial. Terpenes and ashes have to be extracted before the modification because otherwise these components would be washed out during the modification process, which would falsify the observed weight difference of the treated samples. The extraction after the modification in turn serves to remove the unbound residues of the applied modification reagent.

At the beginning of the modification, the wood sample was swollen in a nitrogen atmosphere in a mixture of a base and catalyst in anhydrous DMF for 2 h at 50 °C. During this initial process, the pore structures of the wood expand, which enables the molecules to diffuse more easily into the material and react with the enlarged cell surface. Furthermore, a deprotonation of the wooden hydroxy groups already takes place during the swelling process so that the subsequent esterification is facilitated. After addition of the modification reagent (7.0 mmol per gram wood), the wood sample reacted for 24 h at 70 or 120 °C, respectively. The reaction conditions used for the wood modification are summarized in Figure .

Figure 3

Reaction conditions for wood modification applying activated benzamides.

After modification, extraction, and drying of the wood samples, the mass gain and the bound amount of substance were determined. The obtained weight percentage gain (WPG) and quantity of covalently bound organomaterial (QCO)[48] values are summarized in Table while the corresponding formulas for the WPG and QCO calculations are given in the Supporting Information. Due to the fact that the functionalization was carried out several times at each temperature, the value range of the WPG is shown. In addition, the corresponding mean values of the WPG and QCO values as well as the associated standard deviations are given for each modification reagent.

Table 2

Results of Wood Modification Reactions Applying Activated Aromatic Carboxamides

reagent	temperature (°C)	WPG-range (%)	WPGØ (%)	σWPG (%)	QCOØ (mmol g–1)	σQCO (mmol g–1)
1	70	4.6–5.1	4.9	±0.3	0.46	±0.03
	120	10.3–11.7	11.0	±1.0	1.05	±0.09
2	70	7.1–9.5	8.1	±1.2	0.66	±0.10
	120	18.4–28.9	23.5	±4.9	1.91	±0.40
3	70	11.9–12.9	12.4	±0.7	0.89	±0.05
	120	22.2–22.9	22.6	±0.5	1.62	±0.04
4	70	10.0–14.7	12.4	±3.3	0.89	±0.24
	120	25.0–26.8	25.9	±1.3	1.86	±0.09
5	70	4.0–4.3	4.2	±0.2	0.30	±0.01
	120	13.8–14.9	14.3	±0.8	1.03	±0.06
6	70	12.2–17.3	15.0	±2.2	0.82	±0.12
	120	27.5–32.6	30.1	±3.6	1.64	±0.20
7	70	10.5–12.3	11.5	±0.8	0.88	±0.06
	120	11.0–17.0	14.0	±4.2	1.07	±0.33
8	70	15.1–18.0	16.6	±2.0	1.10	±0.14
	120	33.0–33.0	33.0	±0.0	2.20	±0.00
9	70	9.4–15.5	12.4	±4.3	0.83	±0.29
	120	22.0–23.6	22.8	±1.1	1.52	±0.07
10	70	2.5–4.8	3.7	±1.6	0.24	±0.11
	120	10.6–11.4	11.0	±0.6	0.73	±0.04
11	70	0.4–0.6	0.5	±0.1	0.04	±0.01
	120	1.5–1.7	1.6	±0.1	0.14	±0.01
12	70	0.5–0.8	0.7	±0.2	0.05	±0.01
	120	1.9–2.3	2.1	±0.3	0.14	±0.02
13	70	3.3–5.5	4.4	±1.6	0.25	±0.09
	120	15.3–16.5	15.9	±0.9	0.90	±0.05
14	70	–0.4–0.2	-0.3	±0.1	n/a	n/a
	120	–1.0–0.8	-0.9	±0.1	n/a	n/a
15	70	4.1–5.3	4.7	±0.8	0.35	±0.06
	120	9.4–11.1	10.2	±1.2	0.76	±0.09
16	70	–0.2–0.4	0.1	±0.4	0.01	±0.03
	120	1.0–1.7	1.3	±0.5	0.09	±0.03
17	70	9.2–10.0	9.6	±0.6	0.57	±0.04
	120	17.0–20.4	18.7	±2.4	1.11	±0.14
18	70	1.9–2.4	2.1	±0.3	0.13	±0.02
	120	7.1–8.8	8.0	±1.2	0.48	±0.07
19	70	14.7–16.9	15.8	±1.6	0.88	±0.09
	120	27.2–30.2	28.7	±2.2	1.59	±0.12
20	70	7.1–8.5	7.8	±1.0	0.52	±0.07
	120	14.4–14.8	14.6	±0.3	0.97	±0.02
21	70	15.5–21.0	18.3	±3.9	1.00	±0.21
	120	21.2–25.0	23.1	±2.6	1.26	±0.14
22	70	12.7–12.7	12.7	±0.0	0.78	±0.00
	120	20.7–24.6	22.7	±2.8	1.39	±0.17
23	70	3.9–5.3	4.6	±1.0	0.39	±0.08
	120	11.2–12.2	11.7	±0.7	0.98	±0.06
24	70	17.7–18.6	18.1	±0.7	1.05	±0.04
	120	29.1–35.9	32.5	±4.8	1.88	±0.28
25	70	7.9–8.2	8.1	±0.2	0.53	±0.02
	120	11.3–13.1	12.2	±1.3	0.80	±0.08
26	70	3.4–9.4	6.9	±2.5	0.35	±0.13
	120	13.1–17.4	15.2	±3.0	0.77	±0.15
27	70	5.3–7.4	6.3	±1.5	0.35	±0.08
	120	16.7–17.4	17.0	±0.5	0.94	±0.03
28	70	–0.2–0.1	–0.1	±0.0	n/a	n/a
	120	–0.8–0.6	–0.7	±0.2	n/a	n/a

The modified wood chips show significant weight gains, which confirms the successful covalent attachment of the carboxamides to the wood biopolymers. The maximum WPG value of 33% with a corresponding QCO value of 2.20 mmol g–1 was achieved by 4-nitrobenzamide 8 at a reaction temperature of 120 °C. Only 4-hydroxybenzamide 14 and the anthracenyl derivative 28 show a slight decrease in mass. As a result, no QCO values can be calculated for these two modifications. However, Kaldun et al.[47] showed that the applied modification procedure also entails a small loss of lignocellulosic material in the percentage weight range from −0.8 to −1.7%. This comparison gives evidence that the derivatives 14 and 28 were tied to the wood polymers since the weight loss of these samples was smaller than the comparative experiments without the substrate carried out by Kaldun et al.[47] For a comparative evaluation, the QCO values of the benzoylated samples are plotted in descending order according to the values at 70 °C (Figure ) and 120 °C (Figure S1, see the Supporting Information), respectively.

Figure 4

Summary of QCO values—sorted by descending values at 70 °C.

The QCO values show that larger amounts of bound precursors are achieved at 120 °C in contrast to 70 °C. While some carboxamides show up to 90% higher QCO values at 120 °C, other compounds only provide about 20% higher values. This result proves that temperature has a significant influence on the reactivity of the applied carboxamides. In addition, the average standard deviation of the QCO values is larger at higher temperatures, resulting in a lower reproducibility of the modification results at 120 °C. This observation can be explained by the fact that additional, thermally induced side reactions of the biopolymers occur at elevated temperatures. In addition to a cross-linking mechanism of the biopolymers via the elimination of water, there also occurs thermal degradation of the biopolymers, especially the least stable hemicellulose.[1,26,61] The thermally induced change of the wood material during the chemical modification also becomes evident from a darker coloration of the wood sample at higher temperatures (Figure ).

Figure 5

Thermally induced color change of the wood chips modified with reagent 6 at 70 °C (left) and 120 °C (right).

The evaluation of the QCO values allows us to draw a conclusion about the reactivities of the applied benzamides. The derivatives substituted with electron-withdrawing groups (4-NO2, 4-CF3, 4-SO2Me, 4-halo, 4-CN) are particularly reactive since the electron density in the aromatic ring is reduced and consequently, the carbonyl carbon atom is activated for nucleophilic attacks. In contrast, the benzamides substituted with electron-donating groups (4-OH, 4-NH2, 4-NMe2) react more slowly with the biopolymers of wood, since an increase of the electron density in the aromatic ring causes deactivation of the carbonyl group. In addition, the steric requirements of the substituents also have an influence on the reactivity during the wood modification. Compounds with a carbonyl group sterically hindered by further substituents (9-anthracenyl carboxylate, 2-NO2) therefore led to very low QCO values.

The results demonstrate a clear influence of the substituents attached to the phenyl ring on the reactivity of the carboxamides in the modification reaction. The strength of the respective electronic effects in meta and para positions can be estimated using the substituent constant σ, which can be calculated from the Hammett equation (Formula ).[62−64]Formula : Hammett equation with k the rate coefficient of the substituted compound, k0 the rate coefficient of the unsubstituted compound, ρ the reaction constant, and σ the substituent constant.[62−64]

The Hammett equation describes a quantitative relationship between the structure of a doubly substituted aromatic system and its reactivity in a selected reaction, such as the hydrolysis of an ester group.[62−64] The proportionality constant ρ is characteristic for the investigated reaction, whereas the substituent constant σ describes the influence of the functional group bound in the meta or para position. The σ values, summarized by Jaffé, are compared with the QCO values of selected wood modifications in a comparison plot in Figure .[64] The entries in the diagram are sorted according to descending σ values (all values are additionally summarized in Table S2 in the Supporting Information). The absolute values must not be compared, only the trend of the reactivities, expressed by the QCO and σ values, allows an interpretation.

Figure 6

Plotting the QCO values and the σ values from the Hammett equation (taken from Jaffé[64]) as comparative reactivity criteria. The entries are sorted according to descending σ values.

A small σ value theoretically corresponds to a low reactivity at the carbonyl carbon atom for a nucleophilic attack by the wood hydroxy groups, which results in a low QCO value of the modification. This theory is confirmed by the approximately parallel progression of the QCO values and the σ values in Figure . The deviations from the reactivity trend are significantly larger in the modifications at 120 °C than in the results obtained at 70 °C. This observation indicates that, in addition to the substituent influences described by the σ values, other factors also affect the modification process. Possible influences like thermally induced side reactions or the steric hindrance of the substituents have already been discussed above. The stability of the substrates and their solubility also influences the result of each wood modification.

In addition to the para-substituted benzamides, ortho-, meta-, or poly-substituted benzamides were also used for the wood modification. While the meta-substituted compounds 4 and 9 (3-Cl, 3-NO2) achieve similar or slightly lower QCO values than the para-substituted derivatives 3 and 8 (4-Cl, 4-NO2), the ortho-substituted compounds 5 and 10 (2-Cl, 2-NO2) lead to significantly lower QCO values. The lower amounts of the covalently bound substance can be explained by the increased steric hindrance of the carbonyl group due to the substituents in the ortho position. This effect is particularly evident with the 2-nitro compound 10 since the nitro group is more voluminous and therefore leads to a lower QCO value than the corresponding 2-chloro compound 5, regardless of the stronger electron-withdrawing effect of the nitro moiety.

The polysubstituted carboxamides are equipped with hydroxy (16 and 18) or methoxy groups (17 and 19) in the para position and are additionally substituted with an EWG in the meta position. These derivatives serve as example compounds since it is possible to replace the proton or the methyl group with other functional units such as insecticides, as shown by Söftje et al.[51]

Because of the fact that the hydroxy- (14) and alkoxy-substituted derivatives (15) only reach low QCO values due to their positive mesomeric effect (+M effect) of the substituents, additional EWGs (NO2 or Cl groups) were attached in the meta position (compounds 16–19). The effect of the additionally attached EWGs on the reactivity becomes evident from the resulting QCO values plotted in Figure S2 (see the Supporting Information).

While compound 14 substituted exclusively with a hydroxy group in the para position indicates negative WPG values, the 3-chloro-4-hydroxy derivative 16 already leads to positive WPG values and thus the amount of the bound substance between 0.01 mmol g–1 (70 °C) and 0.09 mmol g–1 (120 °C). Finally, the 3-nitro-4-hydroxy compound 18 leads to significantly larger QCO values of 0.13 mmol g–1 (70 °C) and 0.48 mmol g–1 (120 °C). A similar reactivity trend can be observed for the methoxy compounds 15, 17, and 19. Although all of these compounds show larger QCO values than the comparable hydroxy derivatives (up to 1.59 mmol g–1 for 19 at 120 °C), a similar reactivity trend can be observed, whereby the influence of the EWG in the meta position becomes evident. In summary, it can be stated that these doubly substituted compounds are well suited as linking moieties to bind functional building blocks with an influence on the wood material properties to the biomaterial.

To prove that the modification method developed by Kaufmann and co-workers[46,48−51] can also be applied to other types of wood such as European beech (Fagus sylvatica L.) and Sycamore maple (Acer pseudoplatanus L.), the corresponding specimens were modified. In contrast to pine wood, these types of wood do not provide any usable sapwood parts, so heartwood veneers were applied for modification. 4-Fluoro- (2) and 4-bromobenzamide (6) were selected as reference substrates and used under the standard modification conditions specified in Figure . The results of the wood modifications are summarized in Table .

Table 3

Results of the Wood Modification of Various Types of Wood Applying Carboxamides 2 and 6

reagent	type of wood	temperature (°C)	WPG-range (%)	WPGØ (%)	σWPG (%)	QCOØ (mmol g–1)	σQCO (mmol g–1)
2 (4-F)	pine	70	7.1–9.5	8.1	±1.2	0.66	±0.10
		120	18.4–28.9	23.5	±4.9	1.91	±0.40
	beech	70	6.4–6.8	6.6	±0.3	0.54	±0.02
		120	11.7–25.7	18.7	±9.8	1.52	±0.80
	maple	70	3.8–4.3	4.0	±0.4	0.33	±0.03
		120	14.1–14.2	14.1	±0.1	1.15	±0.01
6 (4-Br)	pine	70	12.2–17.3	15.0	±2.2	0.82	±0.12
		120	27.5–32.6	30.1	±3.6	1.64	±0.20
	beech	70	12.5–12.9	12.7	±0.3	0.69	±0.02
		120	22.0–24.3	23.2	±1.6	1.27	±0.09
	maple	70	17.2–17.3	17.3	±0.0	0.94	±0.00
		120	28.8–29.9	29.4	±0.8	1.61	±0.04

All samples, regardless of the type of wood, show a significant weight gain after treatment with reagents 2 and 6. The beech samples achieved 77−85% of the QCO values of pine wood for both substrates. On the other hand, the maple specimens show a greater variation of the QCO values. While the fluorine derivative 2 only reaches 49−60% of the pine wood values, similar or even slightly higher QCO values are observed applying the bromine compound 6. In a conclusion, the obtained results prove that the wood modification procedure using activated carboxylic acids can be applied to different types of wood. In addition to the treatment of sapwood, the method also allows the modification of heartwood.

Leaching Test

To prove the stability of the covalent bond formed between the benzoate and the wood biopolymers, a standardized leaching test according to DIN EN 84 was carried out.[56] Since European standardized efficacy tests require wood specimens measuring 15 × 25 × 50 mm3, an upscaled modification procedure was carried out to modify the required samples following Soeftje et al.[51] Thereby, five standardized blocks were modified simultaneously in a specially designed glass reactor using 150 mL of anhydrous DMF, anhydrous triethylamine, and DMAP. For the leaching tests, 4-fluoro-substituted benzamide 2 (1.0 mmol per gram wood) was applied as a modification reagent. This derivative was chosen due to its hydrophobizing properties as well as the analytical detectability of the fluorine moiety for subsequent analysis of the leaching solution. To achieve a sufficient penetration depth of the reagents for the larger pine wood blocks, the modification solution was introduced into the reactor by means of a partial vacuum (0.5 bar), followed by the swelling process (2 h at 50 °C) before increasing the pressure by nitrogen addition to atmospheric pressure. The subsequent reaction was carried out analogous to Figure (24 h at 70 °C) yielding the modified samples with a QCO value of 0.40 ± 0.01 mmol per gram wood on average and a corresponding WPG value of 5.0 ± 0.2%. The modified samples were subjected to the leaching test in which the samples were washed in a water bath following the standardized test procedure DIN EN 84.[56] The same test was carried out with five unmodified samples for comparison. At the end of the test, the samples were dried and weighed. The untreated samples showed a weight loss of 3.2 ± 0.5% (WPL), whereas the esterified samples only showed a smaller weight loss of 1.6 ± 0.1%. This standardized test confirms the stability of the covalent bond since the weight loss of the treated samples falls below the natural value of the untreated specimens instead of exceeding it. On the one hand, the lower mass loss shows that no significant amounts of the modification reagent have been leached. On the other hand, the hydrophobization of the wood caused by the fluorine substituent becomes evident.

In addition, the IR spectra of the modified samples showed no changes after the leaching test. The characteristic bands of the covalently bound fluorine substance 2 remained unchanged in the spectrum. Furthermore, the solvent of the collected leaching solution was removed in vacuo and the resulting residue was examined by means of 1H and 19F NMR spectroscopy. The primarily detected signals could be assigned to the cellulose and lignin biopolymers. Additional aromatic signals were detected in the residue of the modified sample, indicating very small amounts of a hydrolyzed, low molecular fluorine component. A weak signal indicating an aryl-bound fluorine atom was also detected in the corresponding 19F NMR spectrum of the sample at −112.0 ppm. However, it was impossible to determine whether the slightly detected fluoro species actually resulted from the hydrolysis of the esterified wood components or whether it was a hydrolyzed residue of unbound benzamide 2, which was eventually not fully extracted after the modification. Nevertheless, the very small detected amounts of the hydrolyzed fluorine compound are negligible. Due to the fact that the chemical modification is still visible in the IR spectra and that no significant amounts of hydrolyzed substances are detected, the covalent ester bond can be classified as largely stable to hydrolysis.

IR Spectroscopic Characterization

The analysis of the modified wood samples was primarily based on ATR-IR spectroscopy, since this nondestructive method has been proven to be a reliable and meaningful method to examine esterified wood.[46−51,53] In addition, the formation of a covalent bond between the applied reagent and the wood biopolymers has already been confirmed by Drafz et al.[22] and Namyslo et al.[65] via 2D-NMR spectroscopy as well as by Ehrhardt et al.[66] and Soeftje et al.[67] via thermal analyses (pyrolysis GC MS). Furthermore, microcomputed tomography-based investigations provided information concerning the penetration depth and the pathways of the modifying reagent into the wood tissue.[53,54]

The ATR-IR spectra of the chemically modified samples using benzamide 1–28 are presented in Figures –9 each in comparison to unmodified Scots pine sapwood. The analytical comparison of unmodified and modified wood reveals significant changes caused by the formation of a covalent bond between the modifying reagent and the biopolymers. While the distinctive bonds caused by the lignin and the polysaccharides have already been fully elucidated in the literature,[68−73] the newly formed ester bond leads to a significant increase of characteristic vibrations A–D in each spectrum.

Figure 7

ATR-IR spectra of unmodified and modified wood (samples 1–10) in comparison (A: νC=O, B: νC=C, arom., C: νC−O, asym., and D: δC-H, arom.).

Figure 9

ATR-IR spectra of unmodified and modified wood (samples 21–28) in comparison (A: νC=O, B: νC=C, arom., C: νC−O, asym., and D: δC−H, arom.).

ATR-IR spectra of unmodified and modified wood (samples 11–20) in comparison (A: νC=O, B: νC=C, arom., C: νC−O, asym., and D: δC-H, arom.).

The carbonyl stretching band at approximately 1700 cm–1 (area A) as well as the corresponding C–O stretching vibrations at 1250 cm–1 (area C) increase, which unambiguously verified the formation of a covalent bond between the respective benzamides 1–28 and the wood biopolymers. In addition, a slight increase in absorption can be observed for the aromatic C=C stretching vibrations at about 1600 cm–1 (area B), which is caused by the introduced aromatic moieties. The aromatic deformation bands were correspondingly detected at about 760 cm–1 (area D). Furthermore, the functional groups attached to the aromatic ring, such as halo, cyano, nitro, silyl, or methoxy substituents, evoke specific stretching and deformation vibrations, which are clearly visible in the IR spectra between 1150–1600 cm–1 (ν) and 500–900 cm–1 (δ), respectively. These additional bands give evidence for the chemical modification of the wood samples with each specific reagent 1–27. Only the sample treated with the anthracenyl derivative 28 does not show any significant differences from the IR spectrum of unmodified pine wood, which can be explained by the very small amounts of the bound substance. Nevertheless, the corresponding samples feature fluorescent properties caused by the large aromatic π-system, proving the covalent attachment of 28 to the biopolymers.

In addition to the treated pine wood specimens, the modified beech and maple wood samples were also examined via IR spectroscopy (Figure ). The spectra of the unmodified beech and maple wood types differ only slightly from the spectrum of the untreated pine wood. The modified samples show the abovementioned C–O and C=C vibrations in areas A–D, independent from the applied type of wood. The wavenumbers and the intensities of the respective bands are very similar, which confirms the covalent attachment of benzamides 2 and 6 to beech and maple wood in addition to the previously discussed pine wood chips.

Figure 10

ATR-IR spectra of unmodified types of wood (pine in black, beech in green, maple in blue) and the specimens modified with benzamides 2 and 6, respectively (A: νC=O, B: νC=C, arom., C: νC−O, asym., and D: δC-H, arom.).

Conclusions

In this study, a wide range of substituted benzoic acids was activated by means of 1H-benzotriazole, and the compounds were isolated, mostly in very good yields and fully characterized. The reactivity of these benzamides toward wood hydroxy groups was tuned by various EWGs and EDGs attached to different positions of the phenyl ring. The synthesized reagents enabled a detailed investigation of their reactivities in the subsequent benzoylation of wood. The chemical modification of Scots pine sapwood veneer chips, carried out at different temperatures, led to peak values in the covalently bound organomaterial of up to 1.10 mmol per g at 70 °C and 2.20 mmol per g at 120 °C. Comparison of the functionalizations showed that the reactivities of the applied benzamides and thus the modification results depend on the electronic effects of the functional groups attached to the aromatic ring. Aromatic compounds substituted with EWGs can be bound particularly well to pine sapwood, while benzamides with EDGs only gave moderate QCO values and in some cases even led to a weight loss of the treated sample. It can be summarized that our study reveals a correlation between the reactivity of the esterification of lignocellulosic biopolymers and the reactivity constants arising from the Hammett equation. These findings enable a prior estimation of the effectiveness of a wood modification approach to develop effective modification substrates in advance. In addition, the temperature dependence of the modification results was determined, since the functionalizations carried out at 70 °C consistently gave very good results with small standard deviations in the WPG and QCO values. Although the modifications carried out at 120 °C mostly led to larger values in the amount of the bound substance, thermal effects on the wood sample were also observed, which both impaired the visual appearance of the material and caused inadequate reproducibility. In conclusion, a temperature of 70 °C proved to be optimal for wood modification with activated carboxylic acids. It was also shown that a second substituent with electron-withdrawing properties increases the effectiveness of the modification. These doubly substituted linking compounds are well suited to bind other functional building blocks to the biomaterial, which are able to influence wood properties.[51] Furthermore, the process for the chemical treatment of pine sapwood was also applied successfully to other types of wood such as beech and maple heartwood. All modified wood samples were analyzed extensively and the attached fragments were detected by IR spectroscopy. The modification process was scaled up to larger specimens (15 × 25 × 50 mm3) and the wood blocks modified with 2 were subjected to a leaching test following DIN EN 84[56] to investigate the durability of the covalent modification. The standardized test explicitly confirmed the durability of the covalent ester bond between the substrate and the biomaterial. This result finally underlines the suitability of the wood modification method developed by Kaufmann and co-workers.[46−51] Furthermore, this study shows the potential of the described functionalization method since it can basically be transferred to all lignin and/or carbohydrate-containing biomacromolecules like natural fibers or smaller biobased chemicals.

Materials and Methods

General Methods

NMR Instrument

1H NMR (600 MHz), 13C NMR (150 MHz), and 15N NMR (61 MHz): Avance III 600 MHz FT-NMR spectrometer (Bruker, Rheinstetten, Germany). 1H NMR (400 MHz) and 13C NMR (100 MHz): Avance 400 FT-NMR spectrometer (Bruker). 1H and 13C NMR spectra were referenced to the residual solvent peak [CDCl3: δ = 7.26 ppm (1H), δ = 77.0 ppm (13C) or DMSO-d6: δ = 2.50 ppm (1H), δ = 39.5 ppm (13C)]. For the 15N NMR spectra, nitromethane (δ = 0.0 ppm) was used as an internal or external standard. In all cases, peak assignments were accomplished by DEPT135-, HSQC-, and HMBC-NMR experiments. Coupling constants J are given in Hertz (Hz). Multiplicities are described using the following abbreviations: s = singlet, bs = broad singlet, d = doublet, dd = doublet of a doublet, ddd = doublet of a doublet of a doublet, tt = triplet of a triplet, q = quartet, and m = multiplet. Primary and tertiary carbon atoms have been marked with a “+”, the secondary with a “–”, and quaternary with a “Cquat” according to the peak orientation in the DEPT135 spectra.

IR Instruments

Bruker “α-T” (Bruker, Bremen, Germany) equipped with a platinum-ATR-module. IR instruments for wood chips: Bruker “Tensor II” equipped with a platinum-ATR-module.

Mass Spectra Instrument

EI mass spectrometry: a Varian 320 MS Triple Quad GC/MS/MS instrument (Varian, Darmstadt, Germany) with a Varian 450-GC usually operating in direct mode (DEP method) using electron impact ionization (70 eV). In the case of chlorinated and brominated compounds, all peak values of molecular ions as well as fragments refer to the isotopes 35Cl and 79Br, respectively. ESI mass spectrometry: an LC-MSD Series 1100 (Agilent/Hewlett Packard, Santa Clara, CA, US). High-resolution ESI mass spectrometry: a Waters Acquity UPLC coupled to a Waters Q-TOF Premier (Waters, Eschborn, Germany) or an LC-System 1260 Infinity II (Agilent Technologies, Santa Clara, CA, US), coupled to a Bruker Impact II (Bruker, Bremen, Germany). High-resolution EI mass spectrometry: a Waters Micromass GCT (Waters, Eschborn, Germany) operating in direct mode. All HRMS results were satisfactory relative to the calculated accurate mass of the molecular ion (±2.3 ppm, R ≈ 10 000).

Melting Points

Differential scanning calorimeter DSC6 (Perkin-Elmer, Waltham, MA, US). The onset temperature of the endothermic peak in the DSC diagram is evaluated to determine the melting point of each compound.

Wood Modification Reactor

Heidolph Synthesis 1 liquid system parallel synthesizer (Heidolph, Schwabach, Germany).

Chemicals, Solvents, and Wood Materials

Dichloromethane (DCM) was dried using an MP5 solvent purification system from Inert Technology (Amesbury, MA, US). Anhydrous N,N-dimethylformamide (DMF) and all other chemicals were used as purchased from Acros GmbH & Co. KG (Karlsruhe, Germany), Sigma-Aldrich Chemie GmbH (Taufkirchen, Germany), TCI Deutschland GmbH (Eschborn, Germany), or Merck KGaA (Darmstadt, Germany). The untreated Scots pine sapwood veneer samples were obtained from the Section of Wood Biology and Wood Products, the Georg-August-University Göttingen (Göttingen, Germany). The Scots pine sapwood blocks were received from the Federal Institute for Materials Research and Testing (BAM, Berlin, Germany), whereas the European beech and Sycamore maple heartwood veneer samples were provided by Danzer Deutschland GmbH (Kesselsdorf, Germany).

Chromatography

Thin-layer chromatography (TLC) was performed on Merck TLC plates (aluminum-based) silica gel 60 F 254. Purification was carried out using column chromatography on silica gel 60 (Merck). Petroleum ether as the eluent had the boiling range of 60–70 °C.

X-ray Structure Analysis

Chloromethyl benzamide 25 was crystallized from a solution of deuterated chloroform. A suitable single crystal of compound 25 was selected under a polarization microscope and mounted in a glass capillary (d = 0.3 mm). The crystal structure was determined by X-ray diffraction analysis using graphite monochromated Mo Kα radiation (0.71073 Å) [T = 223(2) K], whereas the scattering intensities were collected with a single crystal diffractometer (STOE IPDS II). The crystal structure was solved by direct methods using SHELXS-97 and refined using alternating cycles of least-squares refinement against F2 (SHELXL-97). All non-H atoms were located in difference Fourier maps and were refined with anisotropic displacement parameters. The H positions were determined by a final difference Fourier synthesis.[74]

C14H10ClN3O (M = 271.70 g mol–1) was crystallized in the monoclinic space group 21/c (no. 14), lattice parameters a = 9.675(5) Å, b = 7.826(5) Å, c = 16.853(8) Å, β = 105.15(4)°, V = 1231.7(1)Å3, Z = 4, dcalc. = 1.465 g cm–3, and F(000) = 560 using 2353 independent reflections and 213 parameters. R1 = 0.0779, w2 = 0.1911 [I > 2σ(I)], goodness of fit on F2 = 1.049, and residual electron density 0.739 and −0.564e Å–3.

Further details of the crystal structure investigations have been deposited with the Cambridge Crystallographic Data Center, CCDC 2075908. Copies of this information may be obtained free of charge from The Director, CCDC, 12 Union Road, Cambridge, CB2 1EZ, U.K. (Fax: + 44(1223)-336 033; e-mail: fileserv@ccdc.ac.uk or http://www.ccdc.cam.ac.uk).

Organic Precursors

General Procedure for the Activation of Benzoic Acids[46−50]

1.10–1.68 equiv of thionyl chloride was added to a suspension of carboxylic acid (or sodium carboxylate) and 3.10–3.52 equiv of 1H-benzotriazole in anhydrous DCM. The mixture was stirred for 16–30 h at rt. Subsequently, water or a 2 M solution of hydrochloric acid (aq) was added. Variant 1: The mixture was extracted with DCM and the combined organic phases were washed with water, if necessary additionally with a saturated solution of sodium chloride (aq), and dried over magnesium sulfate. After evaporation of the solvent and adsorption on silica gel, the product was purified by column chromatography and dried in vacuo. Variant 2: The mixture was extracted with DCM and the combined organic phases were washed with a 2 M solution of hydrochloric acid (aq) and water and dried over magnesium sulfate. After evaporation of the solvent, the product was dried in vacuo. Variant 3: The precipitate was collected by filtration, washed with water, DCM, and if necessary EE. The product was dried in vacuo.

Chemical Wood Modification Procedures

General Procedure for the Chemical Modification of Wood Veneer Chips[47,49,51]

Each functionalization was carried out in a Heidolph synthesis 1 parallel synthesizer applying 7.0 mmol of the respective wood modifying reagent 1–28 per 1.00 g of wood veneer (Pinus sylvestris L. resp. Fagus sylvatica L. or Acer pseudoplatanus L., approximately 10 × 10 × 0.7 mm3, 0.04–0.07 g). Prior to the wood modification, the sample was subjected to extraction in a Soxhlet apparatus for 24 h. Thereby, a solvent mixture comprising toluene/acetone/methanol in a 4:1:1 ratio was used as the extractant. The thus pretreated wood was dried at 105 °C for 24 h and subsequently applied for the wood modification. The wood specimen was placed in the reaction tube, evacuated, and flushed with nitrogen thrice. Afterwards, 6 mL of anhydrous DMF, triethylamine (2 equiv relating to the modifying reagent), and 4-(dimethylamino)pyridine (DMAP, 10 mol % of the modifying reagent) were added under a nitrogen atmosphere. The sample was allowed to swell in the solution for 2 h at 50 °C before adding the modifying reagent (7.0 mmol/g wood). The wood specimen was heated in the reaction mixture for 24 h at 70 or 120 °C, respectively. After cooling down to rt, the modified wood chip was washed consecutively with THF (50 mL), chloroform (50 mL), and diethyl ether (50 mL). The sample was extracted again for 24 h applying the same conditions as stated above. Finally, the treated wood chip was dried for 24 h at 105 °C before determining its weight.

Procedure for Chemical Modification of Standardized Woodblocks[51]

The simultaneous functionalization of five standardized woodblocks was carried out in a glass reactor, which was closed with a flat flange, applying 1.0 mmol of the wood modifying reagent 2 per 1.00 g of Scots pine sapwood (Pinus sylvestris L., 15 × 25 × 50 mm3). Prior to the wood modification, the samples were subjected to extraction in a Soxhlet apparatus for 3 days. Thereby, a solvent mixture comprising toluene/acetone/methanol in a 4:1:1 ratio was used as an extractant. The thus pretreated wood was dried at 105 °C for 3 days and subsequently applied for the wood modification. The woodblocks were placed in the reactor, subsequently evacuated using a rotary vane pump, and flushed with nitrogen thrice. Thereafter, the pressure was increased to 0.5 bar via nitrogen supply and kept constant with a diaphragm pump. Contemporaneous, a solution of (1H-benzotriazol-1-yl)(4-fluorophenyl)methanone (2, 1.0 mmol/g wood), anhydrous triethylamine (2 equiv relating to the modifying reagent), and 4-(dimethylamino)pyridine (DMAP, 10 mol % of the modifying reagent) in 150 mL of anhydrous DMF was prepared under a nitrogen atmosphere. This solution was transferred into the reaction vessel via a metal cannula and through a septum with the help of the applied vacuum in the reactor. The woodblocks were allowed to swell in the solution for 2 h at 50 °C and 0.5 bar before increasing the pressure via nitrogen supply to atmospheric pressure. A quantity of 30 mL of anhydrous DMF was added through the septum and thereupon the specimens were heated in the reaction mixture for 24 h at 70 °C. After cooling down to rt, the modified woodblocks were washed consecutively with THF (250 mL), chloroform (250 mL), and diethyl ether (250 mL). The samples were extracted again for 24 h applying the same conditions as stated above. Finally, the treated woodblocks were dried for 3 days at 105 °C before determining their weight.

The leaching tests were carried out according to DIN EN 84.[56] The woodblocks modified with 4-fluorobenzamide 2 were initially placed in a test vessel and weighted down to prevent the samples from floating. The blocks were then completely covered with water. A vacuum of 40 kPa was applied for a period of 20 min and then the samples were impregnated in the aq solution for another 2 h. The water in the test vessel was drained and replaced with fresh water (100 mL of water per wood sample with the dimensions 15 × 25 × 50 mm3). The leaching solution was replaced in total nine times within the following 14 days, exchanging only three-quarters of the water volume in each case to ensure that the wood samples were continuously covered with water. The described water replacement steps took place at the end of the 1st and 2nd days, and another seven times in the remaining 12 days, with at least 1 but a maximum of 3 days between the exchanges. Following the test procedure, the samples were first dried in air and then at 105 °C and finally weighed. The solvent of the collected leaching solution was removed in vacuo and the resulting residue was dissolved in a solution of 0.580 g of LiCl in 10 mL of DMSO-d6. The analyte was subsequently examined by means of 1H and 19F NMR spectroscopy.