Vapor Phosphorylation of Cellulose by Phosphorus Trichlo-Ride: Selective Phosphorylation of 6-Hydroxyl Function—The Synthesis of New Antimicrobial Cellulose 6-Phosphate(III)-Copper Complexes

This research is focused on a synthesis of copper-cellulose phosphates antimicrobial complexes. Vapor-phase phosphorylations of cellulose were achieved by exposing microcrystalline cellulose to phosphorus trichloride (PCl3) vapors. The cellulose-O-dichlorophosphines (Cell-O-PCl2) formed were hydrolyzed to cellulose-O-hydrogenphosphate (P(III)) (Cell-O-P(O)(H)(OH)), which, in turn, were converted into corresponding copper(II) complexes (Cell-O-P(O)(H)(OH)∙Cu2+). The analysis of the complexes Cell-O-P(O)(H)(OH)∙Cu2+ covered: scanning electron microscopy (SEM), attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR), atomic absorption spectrometry with flame excitation (FAAS), and bioactivity tests against representative Gram-negative bacteria (Escherichia coli) and Gram-positive bacteria (Staphylococcus aureus). The antimicrobial tests of synthesized Cell-O-P(O)(H)(OH)∙Cu2+ revealed their potential applications as an antibacterial material.


Introduction
Cellulose is an important structural component of the primary cell wall of green plants and it presents the most abundant organic polymer on Earth [1,2]. Many properties of cellulose depend on its chain length, a topology, and a surface state of the fibre [2][3][4].
Another possibility of chemical modification of cellulose presents a phosphorylation [13]. Cellulose phosphates, more precisely named cellulose-O-phosphates (III or V) (synonyms: cellulose p; phosphocellulose, dihydrogen phosphate cellulose, cellulose, phosphate ester; phosphorylated cellulose), formed in the so-called cellulose phosphorylation reaction, have been used for decades e.g., sodium cellulose phosphate, under trade name calcibind in the treatment of calcium metabolism-related diseases, taking advantage of their high ability to bind calcium ions (e.g., [14][15][16]). Figure 1 presents the structures of various types of cellulose phosphoric (III/V) acids and corresponding phosphates (III/V).
Their chemistry has regularly been reviewed since the early decades of the 20th century, when they were first proposed as flame retardants [17,18]. Cellulose phosphorylation has also been applied in manufacture of cotton textiles (improvement of flame resistance, moderation of hydrophility-hydrophobity, etc.), cellulose-based nano-materials, ion adsorbents, and ion exchangers [19][20][21], etc. Their chemistry has regularly been reviewed since the early decades of the 20th century, when they were first proposed as flame retardants [17,18]. Cellulose phosphorylation has also been applied in manufacture of cotton textiles (improvement of flame resistance, moderation of hydrophility-hydrophobity, etc.), cellulose-based nano-materials, ion adsorbents, and ion exchangers [19][20][21], etc.
Chemical modification of cellulose by phosphorylation also enhances its bioactivity (e.g., the treatment of calcium metabolism-related diseases) and it provides new derivatives and biomaterials with specific end uses (e.g., [24][25][26][27][28]). Therefore, the synthetic chemistry of this class of biomaterials has been developed for decades, affording a variety of synthetic procedures leading to cellulose-phosphates, in majority focused on cellulose-phosphates P(V) [13].
Chemical modification of cellulose by phosphorylation also enhances its bioactivity (e.g., the treatment of calcium metabolism-related diseases) and it provides new derivatives and biomaterials with specific end uses (e.g., [24][25][26][27][28]). Therefore, the synthetic chemistry of this class of biomaterials has been developed for decades, affording a variety of synthetic procedures leading to cellulose-phosphates, in majority focused on cellulose-phosphates P(V) [13].
These procedures afforded cellulose phosphates/cellulose phosphoric acids with differential phosphorus content, dependent on the applied conditions. Such phosphorylations occurred gradually step-by-step ( [33,34]).
These procedures afforded cellulose phosphates/cellulose phosphoric acids with differential phosphorus content, dependent on the applied conditions. Such phosphorylations occurred gradually step-by-step (  However, if the phosphorylation is carried out in mild conditions, only accessible hydroxyl groups are esterified; in other words, the cellulose microfibrils are only phosphorylated on the surface, with typical regioselectivity (primary 6-HO groups). Such conditions are fulfilled during vapor phase reactions. In this paper, we reveal our results on vapor phase phosphorylation of cellulose by means of PCl3 (Cell-OH→Cell-O-P(O)(OH)-H).
As a part of our research program directed on biologically active functionalized phos- However, if the phosphorylation is carried out in mild conditions, only accessible hydroxyl groups are esterified; in other words, the cellulose microfibrils are only phosphorylated on the surface, with typical regioselectivity (primary 6-HO groups). Such conditions are fulfilled during vapor phase reactions. In this paper, we reveal our results on vapor phase phosphorylation of cellulose by means of PCl 3 (Cell-OH→Cell-O-P(O)(OH)-H).
As a part of our research program directed on biologically active functionalized phosphonates [35,36] and their polymer hybrids [37,38], we present our results on PCl 3 vapor phase phosphorylation of cellulose to cellulose-O-phosphates (III) (H-phosphonates) and their conversion into corresponding copper complexes (Figure 4)]. However, if the phosphorylation is carried out in mild conditions, only accessible hydroxyl groups are esterified; in other words, the cellulose microfibrils are only phosphorylated on the surface, with typical regioselectivity (primary 6-HO groups). Such conditions are fulfilled during vapor phase reactions. In this paper, we reveal our results on vapor phase phosphorylation of cellulose by means of PCl3 (Cell-OH→Cell-O-P(O)(OH)-H).
As a part of our research program directed on biologically active functionalized phosphonates [35,36] and their polymer hybrids [37,38]
Cellulose hydrogen phosphates (III) Cell-O-P(O)(OH)-H, obtained by vapor phosphorylation of cellulose by means of PCl3 and subsequent work-up with water, further called cellulose phosphates (III), were characterized using 31

Phosphorylation of Cellulose
The phosphorylation reactions of the cellulose in the exposure of phosphorus trichloride (PCl3) were carried out in the set consisting of two glass weighing bottles: the larger
Cellulose hydrogen phosphates (III) Cell-O-P(O)(OH)-H, obtained by vapor phosphorylation of cellulose by means of PCl 3 and subsequent work-up with water, further called cellulose phosphates (III), were characterized using 31

Phosphorylation of Cellulose
The phosphorylation reactions of the cellulose in the exposure of phosphorus trichloride (PCl 3 ) were carried out in the set consisting of two glass weighing bottles: the larger one (D vs. H: 40 mm × 40 mm) and the inner vessel (D vs. H: 20 mm × 20 mm) (the figure of the reaction vessel is given in the Supplementary part)). A 0.05 g portion of cellulose was poured into the inner vessel. Raschig rings were placed in the larger bottle (h = 1 cm), and then PCl 3 (1 mL) was added, followed by placing the inner vessel (with cellulose) into the bottle with PCl 3 , followed by the hole closing with a lid. Figure 5 presents chemical schemes of vapor phosphorylation of cellulose.  The reactions were carried out for up to 72 h, after which the inner liner was removed from the reactor, the contents were flushed with nitrogen, and then placed in a beaker of water (25 mL). After 15 min., cellulose phosphate (P(III) (9 h) suspension is filtered on a Schott-Duran sintered disc filter funnel, washed on the filter with water (5 mL), and then  The reactions were carried out for up to 72 h, after which the inner liner was removed from the reactor, the contents were flushed with nitrogen, and then placed in a beaker of water (25 mL). After 15 min., cellulose phosphate (P(III) (9 h) suspension is filtered on a Schott-Duran sintered disc filter funnel, washed on the filter with water (5 mL), and then transferred into a beaker (100 mL) with methanol (5 mL). The suspension was stirred by 5 min, again filtered on a Schott-Duran sintered disc filter funnel, and then dried in a vacuum desiccator over solid KOH for 24 h.  Table 2) and stirred for 2 h, then the solution was filtered off, rinsed with water, dried to constant weight at 50 • C, and then transferred to a vacuum desiccator over KOH.

Solubility of Cell-O 6 -P(O)(OH)-H
The solubility of the prepared sample would be the useful attribute in further derivatizations or potent applications. Generally, the solubilities of cellulose phosphates present scarcely explored field. Thus, Reid and Mozano [39] claimed that cellulose-O-phosphates cannot withstand the rigorous treatment of 6 N sodium hydroxide, but in ca. 1 N NaOH are solubilized during 1 h reflux temperature [39], but the cellulose triphosphates (DS = 2.9] swell considerably in water, forming a consistent translucent gel according to Granja [40]. Cellulose phosphates, obtained by molten urea-phosphoric (III/V) acids methods, are initially isolated by the dissolution of the reacted mixtures in 1 N aqueous sodium hydroxide and then precipitated with methanol ( [41]). In a procedure described by Suflet [32], this process was repeated three times, in order to re-move the residual reagents.
We assumed that ionic liquid based solvents that were applied for the dissolution of cellulose [42][43][44][45][46][47][48][49] can also be applied for the dissolution of cellulose-O-phosphates. Table 3 provides the results of our investigations on the solubility of Cell-O 6 -P(O)(OH)-H.
In the only paper of Petreus [33], the Cell-O-P(O)(OH)-H sample (prepared by phosphorylation of cellulose in molten urea-phosphorous acid mixture (P = 13.4%; DS = 0.97)) was dissolved in D 2 O and analyzed on a Avance III 400 spectrometer, operating at 161.97 MHz for 31 P nuclei. 31 P-NMR spectrum of this sample showed a set of thirteen peaks, with the main at 2.58 ppm and two doublets at 4.99-5.29 ppm and at 7.38 ppm, which were assigned by authors to P-O-C6, P-O-C2 and P-O-C3, respectively. All of the signals according to the Authors corresponded to monosubstituted phosphorous acid esters of cellulose. Figure 6 presents structures of Cell-O i -P(OH)-H (i = 2, 3, and 6) and representative dialkylphosphates (III) with primary and secondary alkoxyl, and corresponding 31 P-NMR chemical shifts (δ [ppm]).
We used 31 P-NMR solid state analysis because our Cell-O-P(O)(OH)-H sample has exhibited solubility neither in D 2 O nor in representative ionic liquids (e.g., TBAA).
We assumed that, during the phosphorylation in mild conditions (as we applied), the formation of cellulose 6-phosphate(III) (Cell-O-P(O)(OH)-H) will be preferred due to the highest reactivity of 6-hydroxyl group of cellulose [62]. In Figure 7, the 31 P-NMR spectrum of cellulose-O-phosphate (III) (Cell-O-P(O)(OH)-H) only exhibits one signal with chemical shift δ = 5.067 ppm, which we assigned to 6-phosphate(III) of cellulose (Cell-O 6 -P(O)(OH)-H), resulting from mild conditions of applied phosphorylation (see Table 1 for comparison). This signal, in contrary to earlier reports [33,34], we attached to 6-phosphate (III) structure, due to higher accessibility and reactivity of primary hydroxyl group in the phosphorylation [43], and, because of that, branching at the carbinol carbon C-C*(OH)-C of phosphate (C*-O-P(O)(OH)-H) usually affords upfield shifts of the phosphorous nuclei (e.g., diethyl H-phosphonate δ 7. Ppm, whereas di-isopropyl H-phosphonate δ 3.5 ppm) [62] (Table 4). Figure   We used 31 P-NMR solid state analysis because our Cell-O-P(O)(OH)-H sample exhibited solubility neither in D2O nor in representative ionic liquids (e.g., TBAA).
We assumed that, during the phosphorylation in mild conditions (as we applied), formation of cellulose 6-phosphate(III) (Cell-O-P(O)(OH)-H) will be preferred due to highest reactivity of 6-hydroxyl group of cellulose [62]. In Figure Table 1 for comp son). This signal, in contrary to earlier reports [33,34], we attached to 6-phosphate ( structure, due to higher accessibility and reactivity of primary hydroxyl group in phosphorylation [43], and, because of that, branching at the carbinol carbon C-C*(OH of phosphate (C*-O-P(O)(OH)-H) usually affords upfield shifts of the phosphorous nu (e.g., diethyl H-phosphonate δ 7. Ppm, whereas di-isopropyl H-phosphonate δ 3.5 pp [62] (Table 4).

SEM-Scanning Electron Microscopy of Cellulose Phosphates
SEM was employed to evaluate the morphological structures of the cellulose phosphates studied. Table 4 characterizes the morphology of various types of cellulose and their derivatives. Figure 9 presents the SEM images ( ×1000 and ×5000 magnifications) of cellulose sam-

SEM-Scanning Electron Microscopy of Cellulose Phosphates
SEM was employed to evaluate the morphological structures of the cellulose phosphates studied. Table 4 characterizes the morphology of various types of cellulose and their derivatives.

SEM-Scanning Electron Microscopy of Cellulose Phosphates
SEM was employed to evaluate the morphological structures of the cellulose phosphates studied. Table 4 characterizes the morphology of various types of cellulose and their derivatives. Figure 9 presents the SEM images ( ×1000 and ×5000 magnifications) of cellulose sample, phosphorylated derivatives Cell-O 6 The presented micrographs do not exhibit substantial morphological changes that are caused by the successive derivatization of cellulose, namely Cell-OH (Figure 9a  Similarly, the formation of copper complex (Cell-O 6 does not accompany substantial changes of the morphology, presumable for the reasons cited above. A similar phenomenon was described by Keshk [65]. They observed that the microstructures of structurally different compounds, namely: starting cellulose 6-phosphate (DP = 1), cellulose-6-phosphate 2,3-dialdehyde, and corresponding cellulose-6-phosphate 2,3-diimines, analyzed by SEM, did not exhibit significant changes at (1 k× and 5 k× magnifications).

Attenuated Total Reflection Fourier Transform Infrared (ATR-FTIR) Spectroscopy
Mid-infrared and Raman spectroscopy are versatile tools in the characterization of structural modifications of biomolecules, being complementary techniques for their structural analysis [66] in these structural analysis of various cellulose-O-phosphates ( [67] and Tables 5 and 6).
The FT-IR spectroscopy was used in this work for the study of the chemical structures of the fibers after chemical modification. Figures 10 and 11 show ATR-FTIR spectra of: unmodified cellulose; Cell-O 6 [69]. α-D-Glucose-Gluc-OH; Gluc-O 6 -P(O)(OH)2-Glucose-O 6 -phosphate. Vibrations derived from phosphoric(III/V) functions are marked in red.

Cell-O-P(O)(H)(OH) Cell-O-P(O)(OH) 2 Cell-O-P(O) (OH)(O-Ph) Vibration
A comparison of the FTIR spectra revealed that, for Cell-O-P(O)(X)OH, the appearance of a new band, at 2400 cm -1 , was absent in the matter cellulose. There is a rather intense band at 1725 cm -1 that is not present in the spectrum of the original cellulose.

Alkalimetric Titration
Because of shapes of the titration curves of Cell-O 6   A comparison of the FTIR spectra revealed that, for Cell-O-P(O)(X)OH, the appearance of a new band, at 2400 cm −1 , was absent in the matter cellulose. There is a rather intense band at 1725 cm −1 that is not present in the spectrum of the original cellulose.

Alkalimetric Titration
Because of shapes of the titration curves of Cell-O 6 -P(O)(OH)-H and Cell-O 6 -P(O)(OH) 2 , resulted from one-or two-proton dissociation in reaction with hydroxide anion (Figure 12), such titration allows the identification, estimation, or semi-quantification of phosphoric groups in cellulose phosphoric acids (Table 7).
Antibiotics 2021, 10, x 14 of 28 anion (Figure 12), such titration allows the identification, estimation, or semi-quantification of phosphoric groups in cellulose phosphoric acids (Table 7).    We carried out the direct titration of the sample of Cell-O 6 -P(O)(OH)-H, synthesized, in order to confirm the nature of phosphate function introduced into cellulose molecule by phosphorylation. Figure 13 presents the figure of the titration curve.  We carried out the direct titration of the sample of Cell-O 6 -P(O)(OH)-H, synthesized, in order to confirm the nature of phosphate function introduced into cellulose molecule by phosphorylation. Figure 13 presents the figure of the titration curve. Elemental analyses of prepared cellulose-O 6 -phosphate (III) (cellulose-O 6 -phosphoric (III) acids) samples were accomplished while using combustion analysis (Elemental Analysis) and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). Table 8 summarizes the results.  Elemental analyses of prepared cellulose-O 6 -phosphate (III) (cellulose-O 6 -phosphoric (III) acids) samples were accomplished while using combustion analysis (Elemental Analysis) and Inductively Coupled Plasma-Mass Spectrometry (ICP-MS). Table 8 summarizes the results.   The supplemental results on 31 P-NMR ( Figure 6) and alkalimetric titration (Figure 13) confirm the selective monophosphorylation of 6-hydroxyl group of cellulose. The supplemental results on 31 P-NMR ( Figure 6) and alkalimetric titration ( Figure 13) confirm the selective monophosphorylation of 6-hydroxyl group of cellulose.

Digestion of Samples Prior to Phosphorus and/or Copper Determination
Cell-O 6   The supplemental results on 31 P-NMR ( Figure 6) and alkalimetric titration (Figure 13) confirm the selective monophosphorylation of 6-hydroxyl group of cellulose.

Digestion of Samples Prior to Phosphorus and/or Copper Determination
Cell-O 6
The specific surface area of the unmodified cellulose (Avicel) is equal to 1,99 [m 2 /g] ( Table 11).  Is worth to note, that in Oshima paper [77] the specific surface areas of cellulose adsorbents determined using the N 2 -BET method were 19.2 m 2 /g for phosphorylated bacterial cellulose (PBC), 2.4 m 2 /g for phosphorylated plant cellulose (PPC), whereas 27.3 m 2 /g for BC, and 1.0 m 2 /g for PC. Therefore

Antibacterial Activity
All of the synthesized cellulosic complexes were tested for their antimicrobial activities, in which Escherichia Coli (Gram-negative bacteria, ATCC11229) and Staphylococcus aureus (Gram-positive bacteria, ATCC 6538) were adopted as the bacterium models. Their antibacterial activities were determined with the agar plate diffusion method. Table 11 lists the results of antibacterial activity tests and Figures 17 and 18

Antibacterial Activity
All of the synthesized cellulosic complexes were tested for their antimicrobial activities, in which Escherichia Coli (Gram-negative bacteria, ATCC11229) and Staphylococcus aureus (Gram-positive bacteria, ATCC 6538) were adopted as the bacterium models. Their antibacterial activities were determined with the agar plate diffusion method. Table 11 lists the results of antibacterial activity tests and Figures 17 and 18 [81][82][83][84].
Lower ZOI values of the composites Cell-O 6 -P(O)(O − )-H × Cu 2+ in comparison with ZOI of soluble copper salts/nanoparticles is caused by a strong binding of copper ions by the functionalities of Cell-O 6 -P(O)(OH)-H, namely by hydrogen-phosphate (III) function, and also by surrounding cellulose hydroxyls. This results in a slow release of copper from the surface of composite, presumably driven by a hydrolysis [85][86][87], which limits a concentration of unbounded Cu (II) cations ( Figure 19).         [81][82][83][84].
Lower ZOI values of the composites Cell-O 6 -P(O)(O -)-H × Cu 2+ in comparison with ZOI of soluble copper salts/nanoparticles is caused by a strong binding of copper ions by the functionalities of Cell-O 6 -P(O)(OH)-H, namely by hydrogen-phosphate (III) function, and also by surrounding cellulose hydroxyls. This results in a slow release of copper from the surface of composite, presumably driven by a hydrolysis [85][86][87], which limits a concentration of unbounded Cu (II) cations ( Figure 19).  (Table 11, Figures 17 and 18) [78]. Table 13 lists the reagents and standard solutions applied. All of these materials and solvents were used as received without further purification and were purchased from Merck (Darmstadt, Germany). Double distilled water was used in all of the experiments. Bacterial strains: Escherichia coli (ATCC 25922) and Staphylococcus aureus (ATCC 6538) were purchased from Microbiologics (St. Cloud, MA, USA).

Specific Surface Area
The specific surface area of the investigated samples was measured using the Autosorb-1 (Quantachrome Instruments, Boynton Beach, FL, USA) apparatus. The analysis was performed while using the physisorption method with nitrogen being used as a sorption agent [72]. The measurements were carried out at 77 K. For each experiment, about 1 g of a given sample was weighed and used. Prior to the analysis, the samples were dried in 105 • C for 24 h and then degassed overnight at room temperature.
The five-point Brunauer-Emmett-Teller (BET) method was applied in order to determine the specific surface area. The specific surface area was calculated twice for each sample, using the five-point adsorption isotherm (P/P 0 in the range of 0.10-0.30) and the 39-point adsorption-desorption isotherm.

SEM/EDS-Scanning Electron Microscopy/Energy-Dispersive X-ray Spectroscopy
The microscopic analysis of samples was performed on a Tescan Vega 3 scanning electron microscope (Brno, Czech Republic) with the EDS Oxford Instruments (Abingdon, UK) X-ray micro analyzer. SEM microscopic examination of the surface topography was performed under high vacuum using the 20 ekV probe beam energy. The surface of each preparation was sprayed with a conductive substance (gold), while using a vacuum dust extractor (Quorum Technologies Ltd., Lewes, UK). The magnification was from 500× to 20000×.

ATR-FTIR-Attenuated Total Reflection Fourier Transform Infrared Spectroscopy
The chemical structure of cellulose samples surface was assessed using ATR-FTIR spectroscopy in the range of 400-4000 cm −1 using a spectrometer Jasco's 4200 (Tokyo, Japan) with an ATR attachment Pike Gladi ATR (Cottonwood, AZ, USA).

Potentiometric Titration of Cell-O 6 -P(O)(OH)-H
Potentiometric titrations were performed using a Cerko-Lab System (Gdynia, Poland) microtitrator that was equipped with a combined glass electrode Hydromet ERH-13-6 (Gliwice, Poland). Cell-O 6 -P(O)(OH)-H (5 mg) samples were placed into glass vessel, followed by an addition of water (2 mL). Subsequently, under intensive stirring, the suspensions were titrated with KOH (0.016 M, carbonate-free), under inert atmosphere (Ar bubbling), at room temperature in the pH range of 2-12. Each titration was repeated at least four times.

ICP-MS-Inductively Coupled Plasma Mass Spectrometry-Determination of Phosphorus by Means of Inductively Coupled Plasma Mass Spectrometry
The method consists of the degradation of cellulose-O-phosphate (P(III)) to phosphoric acid (P(V)) ( Figure 15) and the subsequent analysis of the obtained solution using the ICP-MS technique. Degradation/digestion of the sample was carried out in the mixture: nitric acid, hydrogen peroxide, water, and accelerated by ultrasound irradiation (temperature 200 • C, microwave digestion, 15 min.).
The decomposition of samples was carried out in a computer-controlled, closed, singlemodule microwave mineralizer Magnum II (Ertec, Wrocław, Poland), which was equipped with an integrated pressure temperature control. The process was performed by the wet method, in a closed single-module vessel with a 110 mL reaction chamber under elevated pressure. Microwave energy accelerated the degradation processes. The microwaves were absorbed by the reagents (usually acid or salt solutions) resulting in an increase of temperature and pressure, so that the mushroom-shaped membrane rose, and five heads appeared to accelerate the rapid decomposition of the sample or its chemical synthesis.
Elemental analyses (C and H) were recorded on an Elemental Analyzer Euro EA (Eurovector, Pavia, Italy), phosphorus determinations were performed after prior digestion of cellulose phosphate samples, while using an Agilent 7900 ICP-MS Spectrometer (Santa Clara, CA, USA) that was equipped with a quadruple mass analyzer.
The total copper content of the sample M [mg/kg; ppm] was calculated according to the formula [71]: where: C-metal concentration in the tested solution [mg/L]; m-mass of the mineralized sample [g]; and V-volume of the sample solution [mL].
The antibacterial activity of samples was tested by the agar diffusion method using Muller-Hinton medium agar. The test was initiated by pouring each agar onto sterilized Petri dishes and it was allowed to solidify. The surfaces of agar media were inoculated by overnight broth cultures of bacteria (ATCC 25922: 1.2 × 10 8 CFU/mL, ATCC 6538: 1.7 × 10 8 CFU/mL). Samples of the cellulose: phosphorylated derivatives and cellulose Cu-complex (Cell-O 6 -P(O)(OH)-H(48 h) × Cu 2+ ) were placed onto the inoculated agar and then incubated at 37 • C for 24 h. The diameter of the clear zone around the sample was measured as an indication of inhibition of the microbial species. All of the tests were carried out in duplicate. Simultaneously, the same tests were carried out for control samples-samples of unmodified cellulose.