Comparative Studies on Nanocellulose as a Bio-Based Consolidating Agent for Ancient Wood

Abstract

In this work, nanocellulose aqueous dispersions were studied as a bio-inspired consolidating agent for the recovery and conservation of ancient wood and compared with two of the most used traditional consolidants: the synthetic resins Paraloid B-72 and Regalrez 1126. The morphology of crystalline nanocellulose (CNC), determined using Scanning Electron Microscopy (SEM), presents with a rod-like shape, with a size ranging between 15 and 30 nm in width. Chemical characterization performed using the Fourier-Transform Infrared Spectroscopy (FT-IR) technique provides information on surface modifications, in this case, demonstrating the presence of only the characteristic peaks of nanocellulose. Moreover, conductometric, pH, and dry matter measurements were carried out, showing also in this case values perfectly conforming to what is found in the literature. The treated wood samples were observed under an optical microscope in reflected light and under a scanning electron microscope to determine, respectively, the damage caused by xylophages and the morphology of the treated surfaces. The images acquired show the greater similarity of the surfaces treated with nanocellulose to untreated wood, compared with other consolidating agents. Finally, a colorimetric analysis of these samples was also carried out before and after a first consolidation treatment, and after a second treatment carried out on the same samples three years later. The samples treated with CNC appeared very homogeneous and uniform, without alterations in their final color appearance, compared to other traditional synthetic products.

1. Introduction

Recently, one of the most important tasks of nanotechnology has been to develop new materials with a high added value and low environmental impact, starting with renewable and green resources [1,2,3]. In this regard, nanocellulose, thanks to its enormous availability from various natural primary sources and its numerous applications, has turned out to be one of the most interesting innovative materials to enable this technological revolution [4]. Through the mechanical or chemical treatment of wood pulp or vegetable cellulose, it is possible to obtain both cellulose nanocrystals and cellulose nanofibrils with different degrees of crystallinity [5,6,7]. In particular, crystalline nanocellulose (CNC), consisting of cylindrical, elongated, less-flexible, and rod-like nanoparticles 4–70 nm in width, 100–6000 nm in length, and with a 54–88% crystallinity index, is usually produced by acid hydrolysis [4]. The numerous properties of cellulose nanocrystals, such as their nanometric scale, non-toxicity, light weight (1.5 g/cm³), high specific surface area, easy processing, high elastic modulus (140–150 GPa), high aspect ratio and stiffness, gas impermeability, interesting mechanical characteristics, and good thermal stability, suggest the utility of developing novel nanocellulose-based systems [8,9,10]. As well as its other applications, nanocellulose is also used in the conservation and recovery of Cultural Heritage [11,12,13,14,15]. In fact, the collections of objects and artifacts that belong to our artistic patrimony are largely made of cellulosic-based materials, such as paper, painting canvas, and wood. Very often, all these cellulosic materials require consolidation and restoration interventions due to the processes of decay and aging, mainly caused by acidity, which originates from primers, paints, glues, and the absorption of acidic gases from the environment, as well as from the poor conservation conditions to which they are subjected. Usually, the degradation of cellulose-based works of art occurs by the acid-catalyzed hydrolysis of β-1,4-glycosidic bonds, resulting in the loss of their original mechanical properties. This acidification mechanism can be caused by the same materials used in the past for recovery and conservation. For all these reasons, many times, cellulose-based artistic materials require synergistic reinforcement and deacidification interventions [16,17]. In addition, cellulose-based materials can also be subject to biological attacks due to the action of insects and microorganisms that carry out excavator activity [18]. Usually, to preserve a lignocellulosic artefact, chemical consolidants are used; among these, we find the synthetic resins Paraloid B-72 and Regalrez 1126, which can also be used, for example, with the addition of nanoparticles of ZnO for the consolidation of degraded fir wood, thus increasing its resin retention values [12]. Paraloid B-72 is a copolymer of ethyl methacrylate–methyl acrylate that is soluble in various types of solvents—for example, ethanol, acetone, and butyl acetate—that can also be defined as an acrylic resin, with excellent characteristics of hardness, brilliance, and adhesion on a variety of substrates. It is used in the field of restoration as a final varnish or as a consolidator of works of art in wood, marble, stone, metal, etc., at low concentrations (2–4%), and rarely as an adhesive in higher concentrations. Regalrez 1126 is a low molecular-weight aliphatic resin characterized by high resistance to aging and optical properties close to those of natural resins; it is ideal as a mild consolidating agent for wooden artifacts. It is soluble in medium and low-polarity solvents (white spirit, petroleum essence, butyl acetate) and insoluble in water and polar solvents. Another usual approach is based on the use of colloidal systems, as reported by Chelazzi et al. (2017) [19]; alternatively, even the addition of alkaline nanoparticles in these products, such as calcium or magnesium hydroxide, provides an effective action for the deacidification of paper and canvas, as shown by Giorgi et al. (2002), or for the deacidification of waterlogged wood, as described by Poggi et al. (2016) [20,21]. However, these types of chemical products very often do not protect cellulosic materials from aging processes and introduce a lot of problems with toxicity, being harmful to human health [22,23]. In this context, CNC represents the optimal solution for overcoming these problems since, thanks to its affinity with the initial degraded materials, it certainly gives better results compared to synthetic products, and moreover, it is not toxic and does not alter the visual aspect of artistic surfaces [24]. The fields of application for this innovative material and its composite derivatives are numerous [11], such as the deacidification and consolidation of strongly degraded cellulosic artworks [16], the recovery and consolidation of decayed wood [25] and waterlogged wooden artifacts [15], and structural reinforcement for canvases degraded by aging [17]. For example, recent studies by Camargos et al. (2022) demonstrated the protective capacity of green nanocomposites obtained by combining cellulose nanofibrils (CNF), cellulose nanocrystals (CNC), and lignin nanoparticles [26]. Other studies on the use of nanocellulose as a filler for nanocomposite materials have been conducted by Younis et al. (2023), which demonstrated the improvement of mechanical strength in wood artifacts coated with a Klucel E/CNC film. More recently, the same research team (Younis et al., 2024) showed the possibility of using cellulose derivatives, in particular ethers and esters, for the synthesis of nanocomposites useful for the treatment of archaeological wood [27,28].

This work aimed to evaluate the efficacy of CNC in the consolidation treatment of degraded wood compared with traditional products such as acrylic, aliphatic, and aromatic resins, which often alter the visual aspect of artistic surfaces.

On the contrary, nanocellulose, with its excellent properties and its non-toxicity, seems to be a promising candidate as a bio-inspired consolidating agent for wood, with better yields in terms of its stiffness and homogeneity with the starting material [25]. First of all, in this work, CNC was selected as a renewable bio-product, and it was applied as a reinforcing and protective agent on decayed wooden sheet samples with a thickness of 1–2 mm, and was strongly perforated by woodworm [25,29].

2. Materials and Methods

2.1. Materials

All reagents and solvents were of analytical grade and were purchased from Merck. A Milli-Q Element System (Millipore, Molsheim, France) was used to obtain high-purity water for sample preparation and all dilutions. The CNC was prepared following a procedure reported in the literature, with partial modification [25]. Briefly, cotton pulp was hydrolyzed with sulfuric acid (64% w/v) to separate the nanocrystalline phase. An acid/cellulose ratio of 8.75 (w/w) was chosen. The reaction was carried out for 30 min at 60 °C, and then, the hydrolysis was quenched with iced water. The obtained suspension was centrifuged at 6000 rpm for 10 min to eliminate excess water and acid. The precipitate was suspended in distilled water and centrifuged until an opalescent suspension was obtained. The purification of the CNC suspension was carried out using dialysis membrane tubes (cut-off 10–12 kDa, Sigma-Aldrich, St. Louis, MO, USA) in ultrapure water, changing the water every 24 h. The final suspension pH was adjusted to 7 by a small amount of NaOH solution.

Wood samples were obtained from a beam from the 16th century that had been mainly degraded by parasitic attack, taken from the Cabinet of Scientific Research for the Cultural Heritage of Vatican Museums. The pieces of wood (Populus spp.) were cut to obtain a size of 20 × 2 × 80 mm³ (width × thickness × length). Each consolidating product was used to treat the three wood samples, and an untreated sample was taken as a reference. Therefore, the nine treated samples were divided into three groups: A, B, C, corresponding, respectively, to the samples treated with CNC (A1, A2, A3), Paraloid B-72 (B1, B2, B3), and Regalrez 1126 (C1, C2, C3). The untreated sample was indicated by the abbreviation NT. The wood samples were treated with about 1 mL of the consolidating mixtures, described below, by brushing with three consecutive applications on each sample.

The consolidating agents used were:

• Aqueous suspension of cellulose nanocrystals (CNC), used as synthesized;
• Paraloid B-72 and Regalrez 1126 were dissolved in 10% acetone and in 15% white spirit, respectively, and applied.

All wood samples were weighed before the treatment, after 19 h, and one month later (

Table 1. Weights of the samples before the application of consolidants, after 19 h, and after one month.

Sample	Before	After 19 h	After 1 Month
A1	1.343 g	1.356 g	1.342 g
A2	1.021 g	1.031 g	1.020 g
A3	1.301 g	1.312 g	1.298 g
B1	1.280 g	1.406 g	1.392 g
B2	1.305 g	1.408 g	1.396 g
B3	0.974 g	1.061 g	1.050 g
C1	0.989 g	1.073 g	1.061 g
C2	1.376 g	1.529 g	1.509 g
C3	1.673 g	1.791 g	1.773 g

).

All applications were carried out by total impregnation under the hood at a temperature of about 23.5 °C and relative humidity value (RH) equal to 32.1%. Once the treatment was carried out, the samples were left in the hood under the previous controlled conditions, pending characterization.

2.2. Scanning Electron Microscopy (SEM) Characterization

Scanning Electron Microscope (SEM) analyses of nanocellulose were obtained by the high-resolution Auriga-Zeiss (Jena, Germany) Field Emission (FE)-SEM microscope. The concentration of the nanocellulose suspension sample for SEM was 0.05 mg/mL. For the collection of SEM images of treated wood samples, a (FE-SEM) ZEISS Ultraplus was used. The Quorum Q150T ES instrument, a versatile sputter coater/turbo evaporator, which uses a physical vapor deposition (PVD) process to coat the samples with chromium, was used to carry out the metallization of the wood samples before the SEM characterization; the final thickness of the chrome coating was set at 40 nm.

2.3. Fourier Transform Infrared Spectroscopy (FTIR)

The infrared spectrum of the CNC was recorded on a Perkin-Elmer (Waltham, MO, USA) Spectrum One FT-IR spectrometer equipped with an ATR system. The dried sample of CNC for the FTIR spectroscopy was prepared by evaporating the aqueous suspension with a Smart Evaporator Spiral Plug.

2.4. Conductometry and pH Evaluation

The conductivity and pH of the aqueous CNC suspension were measured with a Mettler Toledo (Columbus, OH, USA) Multiparameter Seven Excellence instrument.

2.5. Optical Microscopy

A LEICA DMRX optical microscope (Wetzlar, Germany) equipped with a camera with 2.5×–20× objectives was used to acquire images of the treated and untreated wood samples. For the observation of each sample, three small portions representative of all the material were collected. The images were collected with visible (2.5×) and inducted ultraviolet light (5×).

2.6. Colorimetric Analysis

To verify any color changes induced by the treatments, colorimetric measurements were carried out. The colorimetric measurements of the wood samples were carried out with a KONICA MINOLTA (Tokyo, Japan) integrating sphere spectrophotometer CM-700d model, with illuminating D65 in the CIELAB colorimetric space in SCI (specular component included) and SCE (specular component excluded) mode, with an observer at 10° and a 6 mm spot SAV MINOLTA, (Tokyo, Japan), according to the UNIEN 15886:2010 standard [30]. Each measurement was given by the average of three measurements, repeated three times in the spectral range of 400–700 nm.

3. Results and Discussion

3.1. CNC Characterizations

The electrical conductivity of the CNC suspension was equal to 50.7–50.4 µS/cm; the pH value was equal to 7.07–7.09, as expected.

A morphological and dimensional evaluation of the CNC was carried out using scanning electron microscopy (SEM). A SEM micrograph of the CNC is displayed in Figure 1; the nanocellulose had a rod-like shape with dimensions in the range of 15 to 30 nm in width.

In Figure 2, the FTIR spectrum of the CNC is shown. The spectrum showed an absorbance broad band at 3331 cm⁻¹, given to the stretching vibration of hydroxyl groups (-OH), and a band at 2895 cm⁻¹, related to C-H stretching vibrations. The feature at 1639 cm⁻¹ is associated with -OH bending. Moreover, the peaks at 1429 and 1316 cm⁻¹ are due to the bending vibrations of the C-H and C-O groups present on the polysaccharide rings [31,32,33].

3.2. Characterization of Treated Degraded Wood Samples

In Figure 3, the ten samples are shown: three for each consolidating agent and one untreated, before treatment (Figure 3(1)), immediately after (Figure 3(2)), after 19 h (Figure 3(3)), and 50 days after application (Figure 3(4)).

Immediately after application, some differences between the three consolidating treatments could already be observed: the CNC appeared to have a milky consistency and initially did not penetrate easily into the wood (Figure 4a); moreover, it was observed that this sample had a very high surface tension, which means that the drops, perfectly spherical, remained on the surface and slowly penetrated the wood. Nevertheless, the CNC did not create a film, and therefore did not block subsequent penetrations. On the contrary, the Paraloid B-72 penetrated well from the beginning (Figure 4b); for the Regalrez 1126, this was even possible to see in detail, as shown in Figure 4c. Some spots on the sheet were under the samples, demonstrating that the resin penetrated well into the wood and passed through the holes made by the woodworms.

Table 1 shows the weights of the various samples before the application of consolidants, after 19 h, and one month after the treatment. As can be seen from the data shown in the table, immediately after the consolidating treatment, the samples’ weight increased, and then decreased a little bit due to the drying and evaporation of the solvents in which the consolidants were dissolved. In addition, from this table, it is possible to see that the samples treated with nanocellulose (samples A), one month after treatment, exhibited a slight weight loss compared to their pre-treatment weight, and that this was due to the margin of measurement error. On the contrary, the samples treated with Paraloid B-72 and Regalrez 1126 (samples B and C) increased their weights after one month.

Sample images were obtained in reflected light and were collected with both visible and inducted ultraviolet light at two different magnifications (5× with UV light, Figure 5a,c,e,g; 2.5× with visible light, Figure 5b,d,f,h).

From these images, it is possible to observe the deterioration of the wood. All samples showed signs of biodegradation caused by woodworms throughout the observed area; their presence is demonstrated by the clearly visible spheroidal excrements—for example, in Figure 5a, as indicated by white arrows [34,35].

Moreover, another sign of degradation due to these xylophagous insects is the presence of numerous holes in all the samples, observable in almost all the images in Figure 5—also in this case highlighted with white arrows. As these holes have a very large size, and as the thickness of the samples is very thin, even after the consolidating treatment and regardless of the consolidating solution used they remained empty, as can be seen in the images shown below.

Therefore, from this optical microscopy analysis—and in particular from the UV light images (Figure 5a,c,e,g)—in addition to highlighting the signs of wood degradation, it was also possible to assess the presence of the consolidants applied by the fluorescence, producing an image that was slightly light blue/grey. At the same time, the most important result obtained by the visible light images (Figure 5b,d,f,h) concerns the variation in the visual appearance of the treated samples. In fact, although it was not easy to observe the presence of the different products (CNC (A), Paraloid B-72 (B) and Regalrez 1126 (C)) in these visible light images, it is interesting to see how only the sample treated with CNC turned out to be very similar to the untreated sample in terms of color; on the contrary, the samples treated with Paraloid B-72 and Regalrez 1126 were darker than before the application.

All the colorimetric data collected from the treated wood samples are shown below. Comparing the colorimetric data of the untreated sample with those of the treated samples collected a few days after the first treatment (

Table 2. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with nanocellulose.

		L* (D65)	a* (D65)	b* (D65)
NT	SCI	64.82	9.30	24.03		ΔL	ΔC	ΔE
	SCE	64.71	9.34	24.12	SCI	−0.84	0.32	0.90
Nanocellulose TR	SCI	63.98	9.04	24.21	SCE	−0.85	0.33	0.91
	SCE	63.86	9.08	24.32

,

Table 3. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with Paraloid B-72.

		L* (D65)	a* (D65)	b* (D65)
NT	SCI	64.82	9.30	24.03		ΔL	ΔC	ΔE
	SCE	64.71	9.34	24.12	SCI	−13.58	6.68	15.14
Paraloid B72 TR	SCI	51.24	14.07	28.71	SCE	−13.74	6.81	15.34
	SCE	50.97	14.17	28.92

and

Table 4. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with Regalrez 1126.

		L* (D65)	a* (D65)	b* (D65)
NT	SCI	64.82	9.30	24.03		ΔL	ΔC	ΔE
	SCE	64.71	9.34	24.12	SCI	−12.90	4.13	13.54
Regalrez 1126 TR	SCI	51.92	13.08	25.70	SCE	−13.00	4.17	13.66
	SCE	51.71	13.15	25.81

), it can be observed that the difference in brightness on the L* axis (∆L) for the CNC-treated samples was much lower than the ∆L of the Paraloid B-72 and Regalrez 1126-treated samples; this means that the CNC did not change the original brightness of the wood. The same consideration can be made for the parameter ∆C, which concerns color saturation. For the most important parameter ∆E, which is the total color difference, this becomes a significant piece of data when it is greater than five; in this case study, while for the CNC it was less than one, for the other two consolidating products it was greater than 13, demonstrating that the color of the samples treated with Paraloid B-72 and Regalrez 1126 varied greatly compared to the untreated sample.

The graph in Figure 6 shows the percentage reflectance values, obtained as a function of the incident wavelength for all the treated and untreated samples, acquired in the SCI and SCE mode, confirming what has just been reported above and demonstrating that the samples treated with CNC were lighter than those treated with Paraloid B-72 and Regalrez 1126, which darkened after treatment.

In this graph, the green curves, corresponding to the samples treated with CNC, are very close to the curves referring to the untreated sample (orange curves); this is because the results obtained from the NT and CNC samples were very similar, as seen previously. On the contrary, the samples treated with Paraloid B-72 (pink curves) and Regalrez 1126 (blue curves) had very similar curves, in some points even overlapping, but distant from the descriptive curve of the untreated sample; in fact, the results obtained from these samples showed a significant chromatic change compared to the untreated wood.

The same colorimetric measurements were carried out on the same samples one month after the previous ones; the results obtained are shown in

Table 5. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with nanocellulose, obtained from the colorimetric analysis carried out one month after the treatment.

		L* (D65)	a* (D65)	b* (D65)
NT POST	SCI	63.06	9.41	23.30		ΔL	ΔC	ΔE
	SCE	62.99	9.43	23.38	SCI	−1.29	0.57	1.41
Nanocellulose POST	SCI	61.77	9.06	23.75	SCE	−1.31	0.58	1.43
	SCE	61.68	9.09	23.84

,

Table 6. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with Paraloid B-72, obtained from the colorimetric analysis carried out one month after the treatment.

		L* (D65)	a* (D65)	b* (D65)
NT POST	SCI	63.06	9.41	23.30		ΔL	ΔC	ΔE
	SCE	62.99	9.43	23.38	SCI	−11.18	4.62	12.10
Paraloid B72 POST	SCI	51.88	12.96	26.25	SCE	−11.29	4.66	12.21
	SCE	51.70	13.01	26.36

and

Table 7. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) and the samples treated with Regalrez 1126, obtained from the colorimetric analysis carried out one month after the treatment.

		L* (D65)	a* (D65)	b* (D65)
NT POST	SCI	63.06	9.41	23.30		ΔL	ΔC	ΔE
	SCE	62.99	9.43	23.38	SCI	−10.66	4.92	11.74
Regalrez 1126 POST	SCI	52.40	12.99	26.68	SCE	−10.69	4.94	11.78
	SCE	52.30	13.03	26.75

.

Moreover, a comparison between the previous colorimetric data and those acquired a month later from the same samples exhibited only small variations in ∆L, ∆C, and ∆E in all cases (

Table 8. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT) with data collected one month earlier from the same sample.

		L* (D65)	a* (D65)	b* (D65)
NT PRE	SCI	64.82	9.30	24.03		ΔL	ΔC	ΔE
	SCE	64.71	9.34	24.12	SCI	−1.76	0.74	1.91
NT POST	SCI	63.06	9.41	23.30	SCE	−1.72	0.75	1.88
	SCE	62.99	9.43	23.38

,

Table 9. Comparison of the colorimetric parameters (L*, a*, b*) of the nanocellulose-treated samples with data collected one month earlier from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Nanocellulose PRE	SCI	63.98	9.04	24.21		ΔL	ΔC	ΔE
	SCE	63.86	9.08	24.32	SCI	−2.21	0.47	2.26
Nanocellulose POST	SCI	61.77	9.06	23.75	SCE	−2.18	0.48	2.23
	SCE	61.68	9.09	23.84

,

Table 10. Comparison of the colorimetric parameters (L*, a*, b*) of the samples treated with Paraloid B-72 with data collected one month earlier from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Paraloid B72 PRE	SCI	51.24	14.07	28.71		ΔL	ΔC	ΔE
	SCE	50.97	14.17	28.92	SCI	0.64	2.69	2.77
Paraloid B72 POST	SCI	51.88	12.96	26.25	SCE	0.73	2.81	2.91
	SCE	51.70	13.01	26.36

and

Table 11. Comparison of the colorimetric parameters (L*, a*, b*) of the samples treated with Regalrez 1126 with data collected one month earlier from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Regalrez 1126 PRE	SCI	51.92	13.08	25.70		ΔL	ΔC	ΔE
	SCE	51.71	13.15	25.81	SCI	0.48	0.99	1.10
Regalrez 1126 POST	SCI	52.40	12.99	26.68	SCE	0.59	0.95	1.12
	SCE	52.30	13.03	26.75

).

As it is possible to observe in the corresponding graph (Figure 7), after one month, the samples treated with CNC did not change their color, but remained very close in colorimetric aspect to the untreated samples; on the contrary, the samples treated with Paraloid B-72 and Regalrez 1126 changed color, making the wood darker.

It is important to note that in the graphs, only four curves were distinguished, and not eight as reported in the legend—this is because for each sample, its two curves always overlapped; therefore, it is only possible to see one curve for each consolidating solution.

The same colorimetric measurements were carried out on the same samples three years after the first treatment, and the acquired data were compared with the previous data collected one month after the first treatment, as shown in

Table 12. Comparison of the colorimetric parameters (L*, a*, b*) of the untreated sample (NT 2024) with data collected three years earlier from the same sample (NT 2021).

		L* (D65)	a* (D65)	b* (D65)
NT 2021	SCI	63.06	9.41	23.30		ΔL	ΔC	ΔE
	SCE	62.99	9.43	23.38	SCI	1.88	1.64	2.50
NT 2024	SCI	64.94	9.56	24.93	SCE	2.06	1.86	2.77
	SCE	65.05	9.64	25.22

,

Table 13. Comparison of the colorimetric parameters (L*, a*, b*) of the nanocellulose-treated samples before the second treatment (Nanocellulose 2024) with data collected three years earlier from the same samples one month after the first treatment (Nanocellulose 2021).

		L* (D65)	a* (D65)	b* (D65)
Nanocellulose 2021	SCI	61.77	9.06	23.75		ΔL	ΔC	ΔE
	SCE	61.68	9.09	23.84	SCI	0.63	0.23	0.67
Nanocellulose 2024	SCI	62.40	8.99	23.53	SCE	0.53	0.28	0.60
	SCE	62.21	9.02	23.58

,

Table 14. Comparison of the colorimetric parameters (L*, a*, b*) of samples treated with Paraloid B-72 before the second treatment (Paraloid B-72 2024) with data collected three years earlier from the same sample one month after the first treatment (Paraloid B-72 2021).

		L* (D65)	a* (D65)	b* (D65)
Paraloid B72 2021	SCI	51.88	12.96	26.25		ΔL	ΔC	ΔE
	SCE	51.70	13.01	26.36	SCI	−0.42	5.54	5.56
Paraloid B72 2024	SCI	51.46	15.43	31.21	SCE	−0.38	5.54	5.55
	SCE	51.32	15.47	31.32

and

Table 15. Comparison of the colorimetric parameters (L*, a*, b*) of samples treated with Regalrez 1126 before the second treatment (Regalrez 1126 2024) with data collected three years earlier from the same sample one month after the first treatment (Regalrez 1126 2021).

		L* (D65)	a* (D65)	b* (D65)
Regalrez 1126 2021	SCI	52.40	12.99	26.68		ΔL	ΔC	ΔE
	SCE	52.30	13.03	26.75	SCI	1.25	4.47	4.65
Regalrez 1126 2024	SCI	53.65	15.42	30.43	SCE	1.26	4.45	4.63
	SCE	53.55	15.46	30.49

.

From these data, we can observe that for the untreated sample (NT), the parameters ∆L, ∆C, and ∆E exhibited a small increase, meaning that the NT sample was brighter, more color-saturated, and had little variation in color—that is, it was lighter compared to the color of the same sample three years earlier.

For the samples treated with nanocellulose, the most important parameter ∆E was still less than five, and this had even decreased compared to 2021; on the contrary, for the samples treated with Paraloid B-72 and Regalrez 1126, all the parameters were increased, showing a variation in the visual aspect of the samples. All these results are described in Figure 8, where the comparison between the colorimetric data of the consolidating agents three years after the first treatment is shown. In this graph, we can observe that, in general, the trend of the curves corresponding to the different consolidants is very similar to that found three years before: the samples treated with nanocellulose were still more like the untreated sample, while the blue curves related to the samples treated with Regalrez 1126 were detached from those in pink of the samples treated with Paraloid B-72— thus showing, over time, a behavior closer to that of nanocellulose.

At this point, the wood samples were treated for a second time with the respective consolidating solutions, and the colorimetric measurements were repeated. In

Table 16. Comparison of the colorimetric parameters (L*, a*, b*) of the nanocellulose-treated samples three years after the first treatment with data collected one week after the second treatment from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Nanocellulose PRE 2024	SCI	62.40	8.99	23.53		ΔL	ΔC	ΔE
	SCE	62.21	9.02	23.58	SCI	0.34	2.88	2.90
Nanocellulose POST 2024	SCI	62.75	9.81	26.29	SCE	0.38	2.96	2.98
	SCE	62.59	9.85	26.42

,

Table 17. Comparison of the colorimetric parameters (L*, a*, b*) of samples treated with Paraloid B-72 three years after the first treatment with data collected one week after the second treatment from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Paraloid B72 PRE 2024	SCI	51.46	15.43	31.21		ΔL	ΔC	ΔE
	SCE	51.32	15.47	31.32	SCI	−2.35	1.23	2.65
Paraloid B72 POST 2024	SCI	49.11	15.55	29.99	SCE	−2.25	1.08	2.49
	SCE	49.07	15.64	30.25

and

Table 18. Comparison of the colorimetric parameters (L*, a*, b*) of samples treated with Regalrez 1126 three years after the first treatment with data collected one week after the second treatment from the same samples.

		L* (D65)	a* (D65)	b* (D65)
Regalrez 1126 PRE 2024	SCI	53.65	15.42	30.43		ΔL	ΔC	ΔE
	SCE	53.55	15.46	30.49	SCI	−2.48	1.85	3.09
Regalrez 1126 POST 2024	SCI	51.17	14.57	28.80	SCE	−2.44	1.90	3.09
	SCE	51.12	14.58	28.80

, a comparison of the colorimetric parameters of each treated sample three years after the first treatment, with the data collected one week after the second treatment from the same samples, is shown.

From the graph in Figure 9, we can observe that after the second consolidating treatment, the green curves of samples treated with nanocellulose were closer to those of untreated sample once again, and that the behavior of the samples treated with Regalrez 1126 returned to being closer to that of the samples treated with Paraloid B-72; however, both were far from the untreated sample. These results demonstrate that treatment with CNC did not change the color of the samples, unlike what happens with samples treated with Paraloid B-72 and Regalrez 1126, which appear darker, in accordance with results obtained three years earlier.

Observing the SEM images of the untreated sample (Figure 10), it is possible to see the fibrous and anisotropic structure of the wood; in particular, it is possible to distinguish both a longitudinal section (parallel to the fibers) and a cross section (perpendicular to the fibers), which is the section that provides more information about the material such as the age of a tree and its growth rate. Moreover, it is possible to see the wood fibers, a parenchymatic ray, and a big woodworm hole in the degraded wood (Figure 10b). Figure 10c shows a longitudinal section at a higher magnification, in which it is possible to observe the cell lumina.

Comparing these longitudinal SEM images with those of the treated samples, it is possible to observe that the wood sample treated with CNC (Figure 11(2) and Figure 12(2)) appears to be coated with a homogeneous film well-adhered to the underlying material, much denser than the other films formed by the Paraloid B-72 (Figure 11(3) and Figure 12(3)) or Regalrez 1126 (Figure 11(4) and Figure 12(4)); it mainly covered the holes, penetrating them and filling the channels. Moreover, from the images, it is almost possible to see the brushstroke with which the consolidant was applied, as if it was an amalgamating consolidator, which binds very well with the underlying material.

This is due to the fact that the nanocellulose, being a constituent of wood, is much more similar to it than the other two synthetic products, and therefore binds better. In particular, in Figure 12(2), it is possible to see how the CNC adhered well in the cell lumina, providing a coating that visually appears more adherent compared to the other products, without sharp edges and with less asperity, as instead can be observed above all in the untreated sample (Figure 12(1)), but also a little in the two other types of treatment (Figure 12(3,4)).

On the contrary, the sample treated with Paraloid B-72 (Figure 11(3) and Figure 12(3)) forms a much harder, seemingly more superficial coating, visually very similar to a cement mixture used as a filler in buildings. In fact, the Paraloid B-72 acts as a very strong and compact coating. In this case, from Figure 12(3), it is possible to see that the edges of the fiber channels, despite being covered by the chemical, remained sharp and defined.

Finally, the sample treated with Regalrez 1126 (Figure 11(4) and Figure 12(4)) presents a very jagged and disordered coating, much less homogeneous and compact than Paraloid B-72 (Figure 11(3) and Figure 12(3)). Additionally, in this case, although the appearance of the treatment is more similar to that of the CNC (Figure 11(2) and Figure 12(2)) than Paraloid B-72, the grooves of the fibrous channels of the wood maintain a defined and sharp appearance. In conclusion, Regalrez 1126 seems to give better treatment in terms of filling than Paraloid B-72, placing itself in second place for results after CNC, which instead turns out to be the best treatment product for affinity with wood.

4. Conclusions

In this work, a new bio-inspired consolidating agent for degraded wood, based on an aqueous solution of CNC, was studied. SEM analyses of this material have shown that it appears very homogeneous and uniform once applied to the artistic surface of degraded wood, unlike the traditional synthetic products Paraloid B-72 and Regalrez 1126, which alter the original appearance of wood and its cell lumina structure, providing thicker surface coatings. Moreover, one month after treatment, it was noticed that the CNC did not increase the weight of the wood samples, unlike what happened for Paraloid B-72 and Regalrez 1126. This is certainly an advantage, because it avoids subjecting already degraded wood to mechanical stress due to the weight of the applied consolidants. In addition, the colorimetric analysis of the untreated and treated samples has demonstrated that the sample treated with CNC did not show alterations in its visual aspect, unlike the samples treated with the synthetic resins Paraloid B-72 and Regalrez 1126, which appear much darker than untreated wood.

Finally, the use of a bio-based restoration product, such as nanocellulose, would certainly be an important step for the ecological revolution necessary not only for the protection of cultural heritage, but also for environmental protection and for the health of operators working in the field of fine arts.

The future prospective is to functionalize the CNC to make it not only a filler agent, but a full-fledged binder of the cellulosic fibers of degraded wood [36,37]—thus also obtaining a conservation and restoration intervention for wooden works heavily degraded by atmospheric and biological agents.