The role of the g-C3N4 precursor on the P doping using HCCP as a source of phosphorus

Original Article The role of the g-C 3 N 4 precursor on the P doping using HCCP as a source of phosphorus Vlastimil Mat ejka a,b,* , Radim  Skuta a,b , Kry stof Foniok a , Vlastimil Nov ak c , Daniel Cvejn b,d , Alexandr Martaus b , Monika Michalska a ,Ji  rı ´Pavlovsky ´ a , Petr Praus a,b a Department of Chemistry, Faculty of Materials Science and Technology, VSB - Technical University of Ostrava, 17. Listopadu 2172/15, 708 00, Ostrava-Poruba, Czech Republic b Institute of Environmental Technology, CEET, VSB - Technical University of Ostrava, 17. Listopadu 2172/15, 708 00, Ostrava-Poruba, Czech Republic c Department of Physical Chemistry and Theory of Technological Processes, Faculty of Materials Science and Technology, VSB - Technical University of Ostrava, 17. Listopadu 2172/15, 708 00, Ostrava-Poruba, Czech Republic d ENET Centre, CEET, VSB - Technical University of Ostrava, 17. Listopadu 2172/15, 708 00, Ostrava-Poruba, Czech Republic article info Article history: Received 13 January 2022 Accepted 3 April 2022 Available online 7 April 2022 Keywords: Graphitic carbon nitride Phosphorus doping Hexachlorocyclotriphosphazene Photocatalysis abstract This work describes the doping of graphitic carbon nitride (g-C 3 N 4 ) with phosphorus performed by 2-h heat treatment of a mechanical mixture of g-C 3 N 4 precursor (urea, dicyandiamide, and guanidine hydrochloride) with hexachlorocyclotriphosphazene at 525 C. The amount of fixed phosphorus in the resulting g-C 3 N 4 structure reached approximately 10 wt% in the case of the urea precursor. For the other two precursors, the fixed phosphorus content in the final products was less than 5 wt%. Several experimental techniques (SEM, XRFS, TG, XRD, FTIR, physisorption of nitrogen, UV-VIS DRS, PL spectroscopy, and electrochemical analysis) were used to characterize the prepared samples. The photodegradation activity of the samples was determined by degradation of Rhodamine B under irradiation with visible light (420 nm). In general, the photodegradation activity of the samples was dependent on the phosphorus content. The highest photodegradation activity was obtained for urea-based g-C 3 N 4 doped with the lowest phosphorus content, with a threefold increase in calcination product yield. The mechanism of incorporation of phosphorus into the final g-C 3 N 4 structure was explained as a two-phase process. ©2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). *Corresponding author. E-mail address: [email protected] (V. Mat ejka). Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate/jmrt journal of materials research and technology 2022;18:3319e3335 https://doi.org/10.1016/j.jmrt.2022.04.019 2238-7854/©2022 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction Graphitic carbon nitride is a well-known and widely studied semiconductor [1e4] with various possible applications. For instance, it can be used in photocatalysis [5e7], sensors [8,9], hydrogen production [10e12], batteries, supercapacitors [13,14], and many more [15e18]. Regarding photocatalytic applications, the main advantages of g-C 3 N 4 include chemical, photochemical, and thermal stability, non-toxicity, and most importantly, low cost and ease of preparation. The narrower band gap energy (around 2.7 eV) compared to TiO 2 (3.2 eV) enables g-C 3 N 4 to absorb VIS irradiation and thus significantly increase the efficiency of light-harvesting for applications based on photocatalytic processes. The frequently used procedure for preparing g-C 3 N 4 is the thermal polymerization of a suitable nitrogen-rich precursor heated in a semi-closed crucible, typically at temperatures in the range of 500e600 C. The g-C 3 N 4 precursor that has received the most attention is undoubtedly melamine [19e22]. Other precursors are urea [23e25], dicyandiamide [26,27], guanidine [28,29], and ionic liquids, such as 1-butyl-3methylimidazolium hexafluorophosphate, which are beneficial as phosphorus sources [30]. However, g-C 3 N 4 samples prepared in this way usually suffer from low specific surface area and fast electron-hole recombination, which impairs photocatalytic performance. Therefore, several strategies have been adopted to improve g-C 3 N 4 photocatalytic performance. Thermal, chemical, or mechanical exfoliation processes [31e36] are used to increase the specific surface area of g-C 3 N 4 . Various approaches such as doping g-C 3 N 4 with metallic and nonmetallic elements [37e45], coupling with other semiconductors [46e50], inert matrices [51e53], or a combination of both [54,55] have been used to suppress the electron-hole recombination rate. Recently, the doping of g-C 3 N 4 with non-metal elements has attracted attention, as it allows lowering of the band gap energy and tuning of the valence and conduction bands [40,56]. In addition, doping with phosphorus (P-doping) has been reported to be efficient in improving photocatalytic activity [29,57,58]. Li et al. [59] modified g-C 3 N 4 by phosphorus and boron to weaken van der Waals interactions between individual layers of g-C 3 N 4 . The resulting material, ultrathin codoped g-C 3 N 4 nanosheets, exhibited a high potential for hydrogen production. Regarding P-doping, it is noteworthy to mention frequently used precursors, in particular, inexpensive ammonium hydrogenphosphate [60,61], phytic or phosphoric acids [62,63], 2-aminoethylphosphonic acid [64] and hexachlorocyclotriphosphazene (HCCP) [65,66]. Our group has recently reported in situ P-doping of g-C 3 N 4 using the mixtures of melamine and HCCP [67]. In addition to promising catalytic activities, the resulting materials showed interesting structural differences compared to pristine and other P-doped g-C 3 N 4 . It was demonstrated that the presence of HCCP affects not only the structure of the final material but also the nature of the P-doping. Also, a doping mechanism was proposed, including the indication of possible phosphorus positions in the polymeric structure of g-C 3 N 4 . In this work, we follow up on our previous study [67] and expand the group of g-C 3 N 4 precursors to include urea, dicyandiamide, and guanidine hydrochloride. The samples were prepared by a one-pot thermal polymerization procedure and further characterized using the methods of physicochemical and phase analysis. The photocatalytic performance of the prepared samples was evaluated by the visible light induced photodegradation of Rhodamine B. 2. Materials and methods 2.1. Chemicals All of the chemicals used were purchased from companies listed below in reagent grade, analytical grade, or equivalent quality. Urea (UR) came from Carl Roth Co. GmBH (Germany), dicyandiamide (DCDA) was obtained from Merck KGaA (Germany), guanidine hydrochloride (GUA) was obtained from VWR LifeSciences (Czech Republic), hexachlorocyclotriphosphazene was obtained from Sigma Aldrich (USA), Rhodamine B (RhB) was obtained from Fisher Scientific (USA). The water used for the preparation of all solutions was deionized and prepared using reverse osmosis (Aqua Osmotic, Czech Republic). 2.2. Sample preparation Phosphorus-modified graphitic carbon nitride was prepared according to our previous study [67] by the 2 h long thermal polymerization of a mechanical mixture of the given g-C 3 N 4 (further labeled as CN) precursor (UR, DCDA, and GUA) and HCCP at 525 C with a heating rate of 2 C,min 1 . After this period of heating, the samples were allowed to cool down inside the furnace. The undoped CN was prepared by the same thermal treatment of pure precursors (UR, DCDA, and GUA). For all calcination experiments, the initial mechanical mixture (5 g) containing the given CN precursor with 0, 1, 10, 20, 30, and 40 wt% of HCCP was placed in a semi-closed ceramic crucible and subjected to the heat treatment mentioned above. The sample labelling system is described in Table 1. 2.3. Characterization methods 2.3.1. X-ray powder diffraction X-ray powder diffraction (XRD) patterns were measured using a SmartLab diffractometer (Rigaku, Japan), equipped with a detector D/teX Ultra 250. A cobalt tube (CoKa, a 1 ¼0.178892 nm, a 2 ¼0.179278 nm), operated at 40 mA and Table 1 eDenotation of samples. Temperature Un-doped samples P doped samples UR UR-CN URx-PCN a DCDA DCDA-CN DCDAx-PCN a GUA GUA-CN GUAx-PCN a a x - wt% of HCCP in the pyrolysis mixture (1, 10, 20, 30 and 40). As an example, the sample denoted as UR20-PCN was prepared by the calcination of urea þHCCP mechanical mixture with HCCP content of 20 wt%. journal of materials research and technology 2022;18:3319e33353320 40 kV, was used as a radiation source. The powder samples were ground in an agate mortar, pressed using a microscope glass in a rotational sample holder, and measured in a reflection mode. The samples were rotated at constant speed of 15 rpm during measurement to eliminate the preferred orientation effect. XRD patterns were obtained in the 2 theta range of 5e90with a step size of 0.01and the speed of 0.5min 1 . 2.3.2. Elemental analysis The content of C, N, and H was measured using a combustion method with a CHSN628 elemental analyser (Leco, USA). In a typical experiment, the powder sample (100 mg) was weighed in tin foil cups and transferred to a furnace by the autosampler. 2.3.3. FTIR spectroscopy The Fourier transform infrared (FTIR) spectra of the prepared samples were measured using a Thermo Scientific Nicolet iS10 FTIR spectrometer (Thermo Scientific, USA). Measurements were carried out in the range of 500e4000 cm 1 with a resolution of 2 cm 1 using an attenuated total reflection (ATR) technique. Each spectrum obtained was the average of 64 scans. Before each measurement, the background was scanned to eliminate apparatus and environmental effects. 2.3.4. X-ray fluorescence spectroscopy X-ray fluorescence spectroscopy (XRFS) was performed using an energy dispersive spectrometer Spectro XEPOS equipped with a 50 W Pd X-ray tube. The samples were inserted into a plastic cuvette sealed with a protective Mylar foil and subsequently analysed in the helium atmosphere. 2.3.5. Scanning electron microscopy For microscopic investigations of the samples, a scanning electron microscope (SEM: Tescan Vega) with a tungsten cathode and energy-dispersive X-ray spectroscopy (EDS: EDAX) was used. The SEM micrographs were obtained using a secondary electrons (SE) and backscattered electrons (BSE) mode with an acceleration voltage of 30 kV. Before measurement, the samples were gold sputtered to ensure adequate electron conductivity. 2.3.6. UV-VIS DRS spectroscopy UV-VIS diffusion reflectance spectra were obtained using a Shimadzu UV-2600 spectrometer (Shimadzu, Japan), equipped with an integrating sphere 2600 Plus. Pulverised samples were placed inside a holder, and the spectra were recorded at room temperature in the range of 200e900 nm. Extra pure barium sulphate powder was used for baseline registration. The Tauc plots were used to evaluate indirect band gap energies. 2.3.7. Electrochemical testing Electrochemical testing was used to determine the VB and CB energies using Mott-Schottky approach. The electrochemical measurements were performed in three-electrode system using Autolab PGSTAT302 potentiometer (Metrohm). Glassy carbon electrode (GCE) (Metrohm), Ag/AgCl (3 M KCl) electrode (Metrohm), and Pt sheet (Metrohm) served as a working, reference, and counter electrode, respectively. For each of the experiment, the surface of GCE was coated with thin layer of tested sample by applying of 30 ml of water suspension of tested sample (10 mg of sample in 5 ml of deionized water) on a top of GCE tip. In following step, the coated GCE tip was dried at 85 C for 30 min. The measurements were conducted in 0.1 M KCl electrolyte (pH ¼5.8), and prior the measurement the electrolyte was purged with nitrogen for 30 min. All electrochemical performances were achieved at room temperature. For the measurement an AC signal having an amplitude of 10 mV and frequency 300 Hz was applied. 2.3.8. Photoluminescence spectroscopy Emission PL spectra were acquired on a spectrometer FLSP920 series spectrometer (Edinburgh Instruments, Ltd.) equipped with a Xe900 arc non ozone lamp (450 W) and an R928P type detector (PMT detector). The excitation wavelength of 325 nm was used for all measurements. 2.3.9. Specific surface area The specific surface area (SSA) of the samples was measured at 77 K under less than 1Pa vacuum using a Sorptomatic 1990 device (ThermoFinnigan, USA) with gaseous nitrogen as adsorbate. SSA values were calculated according to Brunauer eEmmett eTeller (BET) theory in the relative pressure range of 0.05e0.25. 2.3.10. Thermal analysis Thermal gravimetry (TG) of the UR, DCDA, GUA and their mechanical mixtures with 20 wt% of HCCP were carried out using a Setaram SETSYS 18TM thermal analyser and an S-type measuring rod. Samples (10 mg) were placed in an alumina crucible covered with an alumina lid. The analysis was carried out in the dynamic atmosphere of argon (the flow rate of 1.5 dm 3 h 1 ) with the heating rate of 10 C,min 1 from 20 Cto 900 C. 2.3.11. Determination of the phosphate ions To determine the presence of nonbonded phosphate ions, the water suspensions of the synthesized samples (0.015 g in 50 cm 3 deionized water) were prepared and mixed for 2 h using an electromagnetic stirrer (250 rpm). After this period, suspensions were vacuum filtered through a membrane filtrate paper with porosity of 0.60 mm (Pragopor, Czech Republic). The concentration of phosphates was determined by VIS spectrometry using an internally validated molybdenum blue method. In our case, 30 cm 3 of a filtrate was taken into a 50 cm 3 volume flask and sulphuric acid (5.4 mol dm 3 ), (NH 4 ) 2 Mo 7 O 24 .4H 2 O and (NH 4 ) 2 Fe(SO 4 ) 2 solutions, each of 5cm 3 volume, were added. After 10 min leaving to stand, the colored solutions were poured into 2 cm optical length cuvettes and visible light absorbance (l¼700 nm) was measured using a Helios bUV-VIS spectrophotometer (UNICAM). 2.3.12. Photodegradation experiment The photocatalytic activity of selected CN and PCN samples was evaluated using the VIS assisted photodegradation of Rhodamine B (RhB) as the model compound in aqueous solution. A LED diode lamp with the maximum light intensity at 420 nm (the light intensity on the irradiated suspension surface was 7 mW cm 2 ) was used for suspension irradiation. journal of materials research and technology 2022;18:3319e3335 3321 In a typical photocatalytic procedure, 0.05 g of the pulverised sample was mixed with 165 cm 3 of the RhB solution with a concentration of 10 mg$dm 3 . The prepared suspension was stirred at 250 rpm in darkness for 90 min to achieve absorption equilibrium. In the next step, the prepared suspension was stirred at 250 rpm under VIS irradiation for 90 min and 4 cm 3 of samples were taken in 15, 30, 60 and 90 min. The suspensions taken were filtered using a 0.45 mm pore size syringe filter (CHROMAFIL GF/RC-20/25 filters, Macherey-Nagel, Germany) to separate the CN material. The absorption spectra of the residual dye were measured using a UV-2600 spectrometer (Shimadzu, JP) in 1 cm quartz glass microcuvettes. To understand RhB photodegradation, photodegradation tests were also performed in the presence of ethylenediaminetetraacetic acid (EDTA), 1,4 benzoquinone (1,4-BQ) and t-butanol (t-Bu) as scavengers of the holes (h þ ), superoxide radicals (O 2 - ) and hydroxyl radicals (OH). 3. Results and discussion 3.1. Yield of calcination The yields of the calcination products for the given precursor and the amount of HCCP are compared in Fig. 1. The highest yields in the range of 42e65% were achieved for the DCDA precursor. Calcination of the mixtures of UR þHCCP and GUA þHCCP led to a significant decrease in the yield of the calcination product, reaching a maximum of 29 and 25% for GUA and UR, respectively. Compared to the previous study on melamine, where the yields ranged from 63% to 43% at 525 C [67], the yields of CN prepared from guanidine and urea CN precursor are significantly lower. Even with such a high mass loss in the case of the UR precursor, visible evaporation of the gaseous products was not observed during heating of the respective mechanical mixtures. Figure 1 also shows that the trends in the dependencies of product yields on HCCP content for GUA and DCDA are similar and show an almost constant value of up to 10 wt% of HCCP followed by its decrease with growing HCCP content. In the case of UR, Fig. 1 shows an increased tendency of the product yield up to 20 wt% of HCCP followed by its decrease with the HCCP content. Our previous research focused on the P-doping of CN prepared by the same calcination procedure applied to the mixture melamine þHCCP revealed a continuous decrease in product yield with the HCCP content [67]. When the trends in the yields are compared, DCDA and GUA behave similarly to UR. The product yield of g-C 3 N 4 prepared from pure UR precursor (UR-CN) is the lowest, reaching only 3%. This indicates that urea is the least thermally stable precursor. The presence of HCCP enhanced the overall yield of the CN synthesis. This effect is evident in the case of urea, when 1 wt % of HCCP in the reaction mixture triples the overall yield, and 10 wt% of HCCP in the reaction mixture increased the yield more than eight times. The content of HCCP up to 10 wt% in the reaction mixture appears to benefit the CN yield. This leads to the hypothesis that HCCP participated in the CNforming reactions as a P-donor and an activating agent for CN precursors. 3.2. Scanning electron microscopy The character of the CN (both pure and P-doped) particles synthesized using UR, DCDA, and GUA as the precursors are shown in Fig. 2 a)-f). The obtained images indicate that the finest morphology was obtained for the sample UR-CN (Fig. 2a-b), while both samples (un-doped and P-doped) prepared from DCDA show a dense microstructure (Fig. 2c-d). Regarding the sample prepared with UR, the agglomeration effect of HCCP on the final sample UR-20HCCP is evident since the morphology of the URCN particles (Fig. 2a) shows a very fine microstructure compared to the particles synthesized using the mixture of URþ20wt% of HCCP (Fig. 2b). The GUA-CN sample shows the most organized microstructure (Fig. 2e), and the doping of this sample with phosphorus caused a slight disintegration of the particles (Fig. 2f). Comparing the images shown in Fig. 2a, c, and e, it is evident that the g-C 3 N 4 precursor (UR, DCDA or GUA) is the main reason for the different morphologies of the final calcination products. 3.3. Elemental composition The comparison of the theoretical amount of phosphorus (P T ) carried with the given amount of HCCP and the real amount of phosphorus determined using the XRFS technique (P content )is shown in Table 2. The results in Table 2 show that the real phosphorus content increased with increasing HCCP content and demonstrate that most of the samples had phosphorus yield higher than 100% (P content /P T >1). This yield shows that the given CN precursor was intensively decomposed, while the HCCP-derived phosphorus remained in the final structure of the calcination products. This is in good agreement with the results presented in Fig. 1a, which showed the lowest yield of calcination products for UR. However, for the samples DCDA1-PCN, DCDA20-PCN, DCDA40-PCN, and GUA40-PCN, the amount of captured phosphorus was lower than the theoretical one. These results also confirm the highest thermal stability of DCDA and the Fig. 1 eYields of calcination based on HCCP content. journal of materials research and technology 2022;18:3319e33353322 lowest thermal stability of UR, considering the same decomposition behavior of HCCP in all mixtures. The overrepresentation of phosphorus in the final materials tends to decrease with the initial content of HCCP in the mixture and by the CN precursor in the row UR >> GUA >DCDA (compare P content /P T in Table 2). The decrease in the P content /P T ratio with an increased concentration of HCCP is most probably attributed to the competition between the reaction of HCCP with the CN precursor and the thermal decay of HCCP. The higher the ratio of HCCP in the mixture, the more decay of HCCP Fig. 2 eSEM images of the samples prepared from urea: a) UR-CN, b) UR20-PCN, c) DCDA-CN, d) DCDA20-PCN, e) GUA-CN, f) GUA20-PCN. Table 2 eComparison of the theoretical phosphorus content (P T ) and phosphorus content in prepared samples (P content ). Sample P T (wt%/mol%) P content (wt%/mol%) P content /P T (wt%/wt%) P content /P T (mol%/mol%) UR1-PCN 0.268/0.065 5.96/1.94 22.24 29.72 UR10-PCN 2.67/0.70 11.3/3.95 4.23 5.64 UR20-PCN 5.35/1.5 12.7/4.40 2.37 2.89 UR40-PCN 10.7/3.68 19.1/7.04 1.19 1.91 DCDA1-PCN 0.268/0.073 0.22/0.07 0.82 0.99 DCDA10-PCN 2.67/0.78 4.65/1.59 1.74 2.04 DCDA20-PCN 5.35/1.69 5.27/1.77 0.99 1.04 DCDA40-PCN 10.7/4.05 6.07/2.05 0.57 0.51 GUA1-PCN 0.268/0.075 0.50/0.16 1.87 2.15 GUA10-PCN 2.67/0.81 3.90/1.30 1.46 1.62 GUA20-PCN 5.35/1.74 7.00/2.39 1.31 1.37 GUA40-PCN 10.7/4.16 8.30/2.79 0.78 0.67 journal of materials research and technology 2022;18:3319e3335 3323 occurs. Similar to the observations of yields (Fig. 1), the P content /P T ratios are obviously higher for the UR samples compared to the GUA and DCDA samples. This indicates that urea is more reactive with the PeCl bond in HCCP than GUA (whose nucleophilicity is altered because of its hydrochloride nature) and DCDA (whose nucleophilicity is altered as a result of the presence of electron withdrawing cyanide moiety). The complete chemical composition of the samples is summarized in Table 3. Carbon, hydrogen, and nitrogen contents were obtained using elemental analysis, phosphorus and chlorine contents were analyzed using XRFS, and oxygen content was calculated as the residue up to 100%. Based on the data obtained from XRFS, the chlorine concentrations in all samples are negligible. Measurable amounts of chlorine were obtained for UR40-PCN and GUA20 (40)-PCN, with the measured Cl content in the samples DCDA20 (40)-PCN being very low (Table 3). It should be noted that in the case of the GUA samples, both HCCP and GUA (guanidine hydrochloride) can be the source of the chlorine. In the GUA20-PCN sample, the molar concentrations of chlorine reached the highest value of 0.11 mol%, which means approximately 1 chlorine atom per 900 atoms of PCN, and its minimal impact on the final structure is expected. This is consistent with our previous work focused on the doping of g-C 3 N 4 with phosphorus via the thermal polymerization of the melamine þHCCP mechanical mixture, in which the complete elimination of chlorine was proved by XRFS and X-ray photoelectron spectroscopy [67]. Table 3 shows that the molar ratios of hydrogen and nitrogen appear to oscillate around a typical value for the given CN precursor (UR eN: 41 mol%, H: 27 mol%; DCDA eN: 44.5 mol%, H: 24.5 mol%; GUA eN: 42.5 mol%, H: 26 mol%). On the other hand, the molar ratio of carbon always decreases with the increased presence of phosphorus (Table 3), which is especially visible in the case of UR-samples (27.0 mol% of carbon in UR-CN vs. 21.8 mol% in UR40-PCN) and substantially less notable in other samples. The concentration of oxygen varies randomly, and samples prepared from precursors with one carbon, such as UR and GUA, appear to have higher oxygen concentrations than samples prepared from DCDA. This observation indicates that at least part of the phosphorus, in dependence on the particular CN precursor, replaces carbon in the CN structure. The contents of phosphorus presented in a form of water leachable phosphates (P P ) and phosphorus fixed in the PeCN structure (P F ) are compared in Table 4. The content of fixed phosphorus was calculated as the difference of P content (Table 2)andP P . The presence of the (PO 4 ) 3anions clearly demonstrates the fact that not all phosphorus introduced by HCCP was firmly incorporated in the P-doped CN structure. All samples except DCDA1-PCN show a higher P F content compared to the P P content, which signalize the most of the phosphorus is tightly bonded to the structure. A notable difference in the P F content can be observed from the data presented in Table 4. The data also show that the P F values grow with the phosphorus content in the case of all samples. The highest values of P F content were obtained in the UR-derived samples (up to 14.6 wt%), as evident in Table 4. This fact is consistent with the previously discussed reactivity between HCCP and urea and the high Table 3 eChemical composition of the prepared samples. Sample C (wt%/mol%) H (wt%/mol%) N (wt%/mol%) P (wt%/mol%) a Cl (wt%/mol%) a O (wt%/mol%) b UR-CN 32.8/27.0 2.88/28.4 58.1/41.0 0.82/0.26 <0.01/<0.01 5.40/3.33 UR1-PCN 31.1/26.1 2.87/29.0 56.3/40.6 5.96/1.94 <0.01/<0.01 3.77/2.38 UR10-PCN 27.2/24.5 2.42/26.2 54.9/42.5 11.3/3.95 <0.01/<0.01 4.18/2.83 UR20-PCN 25.1/22.5 2.62/28.1 52.8/40.5 12.7/4.40 <0.01/<0.01 6.78/4.55 UR40-PCN 22.9/21.8 2.29/26.2 50.9/41.6 19.1/7.04 0.13/0.04 4.68/3.35 DCDA-CN 34.2/29.5 2.32/24.0 60.7/44.8 0.10/0.03 <0.01/<0.01 2.68/1.73 DCDA1-PCN 34.0/29.0 2.44/25.0 60.4/44.1 0.22/0.07 <0.01/<0.01 2.94/1.88 DCDA10-PCN 32.4/28.7 2.25/23.9 59.2/44.9 4.65/1.59 <0.01/<0.01 1.50/0.99 DCDA20-PCN 31.5/27.3 2.51/26.1 59.3/44.0 5.27/1.77 0.03/0.01 1.39/0.90 DCDA40-PCN 30.9/27.0 2.45/25.7 58.8/44.1 6.07/2.05 0.05/0.01 1.73/1.14 GUA-CN 33.6/27.9 2.73/27.2 59.5/42.3 0.03/0.01 0.02/<0.01 4.12/2.57 GUA1-PCN 33.0/27.7 2.66/26.8 58.8/42.3 0.50/0.16 0.01/<0.01 5.03/3.16 GUA10-PCN 31.1/26.9 2.51/26.0 57.3/42.4 3.90/1.30 <0.01/<0.01 5.19/3.36 GUA20-PCN 30.6/27.0 2.41/25.5 57.0/43.2 7.00/2.39 0.33/0.11 2.66/1.76 GUA40-PCN 30.6/26.5 2.64/27.5 57.1/42.4 8.30/2.79 0.25/0.07 1.11/0.72 a Determined by ED-XRFS. b calculated to 100%. Table 4 eThe determined content of soluble (P P ) and fixed phosphorus (P F ). Sample P F (wt%) P P (wt%) Sample P F (wt%) P P (wt%) Sample P F (wt%) P P (wt%) UR1-PCN 5.19 0.77 DCDA1-PCN 0.10 0.12 GUA1-PCN 0.48 0.02 UR10-PCN 9.29 2.01 DCDA10-PCN 3.37 1.28 GUA10-PCN 2.61 1.29 UR20-PCN 9.52 3.18 DCDA20-PCN 3.56 1.71 GUA20-PCN 5.16 1.84 UR40-PCN 14.6 4.53 DCDA40-PCN 3.85 2.22 GUA40-PCN 6.31 1.99 journal of materials research and technology 2022;18:3319e33353324 weight loss of urea during the calcination of the UR þHCCP mixtures (Fig. 1). The dependency of the fixed phosphorus (P F ) molar ratio on the molar ratio of the total amount of phosphorus (P content ) is shown in Fig. 3. Unlike carbon nitrides derived from UR and GUA, the fixed phosphorus content in the DCDA-PCN samples stabilized with the increasing HCCP content in the initial mixture and remained below 4 wt% even at the highest concentrations of HCCP (Table 4 and Fig. 3). It implies that each of the CN precursors appears to have its unique capacity to fix phosphorus from HCCP in its resulting P-doped CN structure, while the rest of the phosphorus remained as weakly bonded in the form of water leachable phosphates. 3.4. Thermal analysis The behavior of the pure CN precursors UR, DCDA and GUA and the mechanical mixtures of UR, DCDA and GUA with HCCP (20 wt%) was studied using thermal analysis. The mass changes of the samples during heating are compared in Fig. 4. The TG curves pictured in Fig. 4 indicate that the mechanical mixtures of CN precursors with HCCP are less thermally stable with an on-set temperature degradation slightly above 100 C. Weight loss in this region is caused by the HCCP decomposition that takes place at more than 100 C as reported by Sun et al. and Osada et al. [68,69], and it is supposed that the initial reaction of CN precursors with HCCP led to the formation of a certain HCCP-CN precursor complex. Both proposed processes probably led to the elimination of HCl connected with the weight loss of the samples. Weight losses connected with the gradual polymerization of pure precursors UR, DCDA, and GUA accompanied by the elimination of NH 3 / H 2 O can be identified for GUA at 300 C, for DCDA at 250 C and for UR slightly above 150 C. The addition of 20 wt% of HCCP always shifted those values to lower temperatures and made polymerization more gradual with the more complex weight loss patterns (Fig. 4). Although the course of the TG curves is different for each mechanical mixture, the temperatures of the final weight loss for the given pure CN precursor and its mechanical mixture with HCCP are similar. The most pronounced decomposition of the samples based on the UR precursor (in Fig. 4 samples UR, URþ20HCCP) finished similarly at around 400 C, for the samples GUA and GUAþ20HCCP at around 720 C and for the samples DCDA and DCDAþ20HCCP at around 750 C. This observation indicates an increase in the thermal stability in the order UR <GUA <DCDA. As shown in Chapter 3.1 focused Fig. 3 eRelation between the total content of phosphorus (xP content ) and the phosphorus captured within the CN-structure (xP F ). Fig. 4 eThermally induced changes of the mass of the mechanical mixtures of studied precursors UR, DCDA, GUA and corresponding mechanical mixtures with HCCP (20 wt%). journal of materials research and technology 2022;18:3319e3335 3325 on the calcination yield, the highest weight loss during the calcination at 525 C (the temperature used for the preparation of the CN samples) was observed for UR, followed by GUA, the significantly higher yield of calcination was observed for DCDA which is in good relation to the results of the TG analysis (Fig. 4). Comparing the TG curves of the UR and URþ20HCCP samples (Fig. 4), the different course of the TG curve (different rate of weight loss) registered for the mechanical mixture URþ20HCCP in two temperature regions, indicate that the process of incorporation of phosphorus influences the reaction mechanism. The complexity of the UR transformation was recently illustrated by Kuntz et al. [70]. The authors proposed the model of UR decomposition including the formation and decomposition of eleven different species, connected with release of 6 gaseous products and comprising of twophase changes. On the other hand, the TG curves of GUAþ20HCCP and DCDAþ20HCCP show similar run with the TG curves of pure GUA and DCDA. 3.5. X-ray diffraction The XRD patterns of CN prepared from UR, DCDA and GUA and PCN materials prepared from these precursors are shown in Fig. 5aec. All registered XRD patterns are dominated by two peaks with maxima at around 15 and 322 Theta (CoKa), which are attributed to the (100) and (002) crystal planes of gC 3 N 4 , respectively [12]. The positions of the maxima of the (002) diffraction line did not change in the case of samples prepared from UR and GUA, as evident in Fig. 5a-c. The position of the maximum of (002) diffraction line of the samples DCDA10-PCN, DCDA20-PCN and DCDA40-PCN is slightly shifted towards higher 2Theta angles (Fig. 5b), which signalizes that the incorporation of the phosphorus in the CN structure caused the decrease of the (002) interlayer space. The intensity of (002) diffraction peaks decreased with the incorporation of phosphorus, indicating the decrease in the thickness of the g-C 3 N 4 particles. Similar observation was published by Li et al. [71] for spherically shaped g-C 3 N 4 coupled with carbon dots. 3.6. FTIR spectroscopy The FTIR spectra of the prepared materials are shown in Fig. 6a-c. The broad peak, appearing in the region of 3500 to 3000 cm 1 , is attributed to the stretching modes of the NeH and OeH bonds of free surface amino groups, adsorbed water, and hydroxyl species [72]. The stretching vibrations of aromatic CeN heterocycles are evidenced with characteristic bands in the region from 1635 cm 1 to 1250 cm 1 [73]. The characteristic peak with maximum at 806 cm 1 corresponds to the breathing vibration of the tri-s-triazine ring system Fig. 5 eXRD patterns of UR (a), DCDA (b) and GUA (c) derived CN and PCN samples. journal of materials research and technology 2022;18:3319e33353326 [74]. Comparing the spectra in Fig. 6, all of the prepared samples can be considered as graphitic carbon nitride materials. The details of the FTIR spectra in the region of 900e1250 cm 1 shown in Fig. 6d did not prove the presence of PeN and PeC bonds in samples derived from DCDA and GUA mainly due to the strong absorption of the CeN heterocycle in this region. On the other hand, the sample UR40-PCN shows enhanced absorption in the 800e1000 cm 1 region and the respective peak could be assigned to the PeN stretching mode, indicating that the phosphorus was incorporated into the structure of CN [75]. Similarly, Li et al. [76] observed the peak at 950 cm 1 that indicates the presence of PeN bonds in P-doped g-C 3 N 4 nanowires arrays prepared from the melamine and phosphoric acid. The chemical analysis of the synthesized samples (Table 3) proved significantly higher amount of phosphorus in the UR-derived samples and thus the presence of PeN bonds can be attributed to the process of the carbon substitution by phosphorus in the sample UR40-PCN. At the same time, the presence of PeC bonds can be excluded from the consideration. Heterocycles containing PeCorC]PeC(nitrogen substitution in the g-C 3 N 4 lattice) show strong absorptions in the regions of 750e600 cm 1 where a set of n(PeC) modes is usually visible [77]. This is not the case for any P-doped materials prepared within this study. Therefore, it can be concluded that there is no support on the FTIR spectra for an occurrence of PeC bonds and the nature of P-incorporation into the structure is via the substitution of carbon. Fig. 6 eFTIR spectra of the prepared CN samples and PCN samples from the mixtures of UR þHCCP, DCDA þHCCP and GUA þHCCP calcined at 525 C. Fig. 7 eDependency of specific surface area values on the amount of HCCP in initial calcination mixtures. journal of materials research and technology 2022;18:3319e3335 3327 nanosheets for decomposition of organic pollutants. Front Chem 2019;7. [43] Li Y-Y, Zhou B-X, Zhang H-W, Huang T, Wang Y-M, Huang W-Q, et al. A hosteguest self-assembly strategy to enhance p-electron densities in ultrathin porous carbon nitride nanocages toward highly efficient hydrogen evolution. Chem Eng J 2022;430:132880. [44] Li B, Fang Q, Si Y, Huang T, Huang W-Q, Hu W, et al. Ultrathin tubular graphitic carbon Nitride-Carbon Dot lateral heterostructures: one-Step synthesis and highly efficient catalytic hydrogen generation. Chem Eng J 2020;397:125470. [45] Li Y-Y, Zhou B-X, Zhang H-W, Ma S-F, Huang W-Q, Peng W, et al. Doping-induced enhancement of crystallinity in polymeric carbon nitride nanosheets to improve their visible-light photocatalytic activity. Nanoscale 2019;11:6876e85. [46] Mat ejka V,  Sihor M, Reli M, Martaus A, Ko cı´ K, Kormunda M, et al. Composites g-C3N4 and BiOIO3 for photocatalytic decomposition of N2O. Mater Sci Semicond Process 2019;100:113e22. [47] Nemiwal M, Zhang TC, Kumar D. Recent progress in g-C3N4, TiO2 and ZnO based photocatalysts for dye degradation: strategies to improve photocatalytic activity. Sci Total Environ 2021;767:144896. [48] Zhu Y, Li E, Zhao H, Shen S, Wang J, Lv Z, et al. Carbon nitride derived carbon and nitrogen Co-doped CdS for stable photocatalytic hydrogen evolution. Surface Interfac 2021;25:101262. [49] Qi KZ, Lv WX, Khan I, Liu SY. Photocatalytic H-2 generation via CoP quantum-dot-modified g-C(3)N(4) synthesized by electroless plating. Chin J Catal 2020;41:114e21. [50] Kesarla MK, Fuentez-Torres MO, Alcudia-Ramos MA, OrtizChi F, Espinosa-Gonz alez CG, Aleman M, et al. Synthesis of gC3N4/N-doped CeO2 composite for photocatalytic degradation of an herbicide. J Mater Res Technol 2019;8:1628e35. [51] Hu X, Shu Z, Guo H, Zhou J, Li T, Tan Y, et al. Metakaolinbased nano-structuring of polymeric carbon nitride and synchronous composite construction for superior photocatalytic H2 evolution. Appl Clay Sci 2020;184:105320. [52] Wang X, Zhao Z, Shu Z, Chen Y, Zhou J, Li T, et al. One-pot synthesis of metakaolin/g-C3N4 composite for improved visible-light photocatalytic H2 evolution. Appl Clay Sci 2018;166:80e7. [53] Wang A, Lee C, Bian H, Li Z, Zhan Y, He J, et al. Synthesis of gC3N4/silica gels for white-light-emitting devices. Part Part Syst Char 2017;34:1600258. [54] Wadhai S, Jadhav Y, Thakur P. Synthesis of metal-free phosphorus doped graphitic carbon nitride-P25 (TiO2) composite: characterization, cyclic voltammetry and photocatalytic hydrogen evolution. Sol Energy Mater Sol Cells 2021;223:110958. [55] Kondo K, Murakami N, Ye C, Tsubota T, Ohno T. Development of highly efficient sulfur-doped TiO2 photocatalysts hybridized with graphitic carbon nitride. Appl Catal B Environ 2013;142:362e7. [56] Qi KZ, Cui N, Zhang MJ, Ma YH, Wang GZ, Zhao Z, et al. Ionic liquid-assisted synthesis of porous boron-doped graphitic carbon nitride for photocatalytic hydrogen production. Chemosphere 2021:272. [57] Chai B, Yan J, Wang C, Ren Z, Zhu Y. Enhanced visible light photocatalytic degradation of Rhodamine B over phosphorus doped graphitic carbon nitride. Appl Surf Sci 2017;391:376e83. [58] Mahvelati-Shamsabadi T, Lee B-K. Photocatalytic H2 evolution and CO2 reduction over phosphorus-doped g-C3N4 nanostructures: electronic, Optical, and Surface properties. Renew Sustain Energy Rev 2020;130:109957. [59] Li B, Si Y, Zhou B-X, Fang Q, Li Y-Y, Huang W-Q, et al. Dopinginduced hydrogen-bond engineering in polymeric carbon nitride to significantly boost the photocatalytic H2 evolution performance. ACS Appl Mater Interfaces 2019;11:17341e9. [60] Bellardita M, Garcı´a-L opez EI, Marcı` G, Krivtsov I, Garcı´a JR, Palmisano L. Selective photocatalytic oxidation of aromatic alcohols in water by using P-doped g-C3N4. Appl Catal B Environ 2018;220:222e33. [61] Hu S, Ma L, You J, Li F, Fan Z, Wang F, et al. A simple and efficient method to prepare a phosphorus modified g-C3N4 visible light photocatalyst. RSC Adv 2014;4:21657e63. [62] Fang H-B, Zhang X-H, Wu J, Li N, Zheng Y-Z, Tao X. Fragmented phosphorus-doped graphitic carbon nitride nanoflakes with broad sub-bandgap absorption for highly efficient visible-light photocatalytic hydrogen evolution. Appl Catal B Environ 2018;225:397e405. [63] Yang H, Zhou Y, Wang Y, Hu S, Wang B, Liao Q, et al. Threedimensional flower-like phosphorus-doped g-C3N4 with a high surface area for visible-light photocatalytic hydrogen evolution. J Mater Chem 2018;6:16485e94. [64] Deng Y, Tang L, Zeng G, Zhu Z, Yan M, Zhou Y, et al. Insight into highly efficient simultaneous photocatalytic removal of Cr(VI) and 2,4-diclorophenol under visible light irradiation by phosphorus doped porous ultrathin g-C3N4 nanosheets from aqueous media: performance and reaction mechanism. Appl Catal B Environ 2017;203:343e54. [65] Lan D-H, Wang H-T, Chen L, Au C-T, Yin S-F. Phosphorousmodified bulk graphitic carbon nitride: facile preparation and application as an acid-base bifunctional and efficient catalyst for CO2 cycloaddition with epoxides. Carbon 2016;100:81e9. [66] Hu C, Hung W-Z, Wang M-S, Lu P-J. Phosphorus and sulfur codoped g-C3N4 as an efficient metal-free photocatalyst. Carbon 2018;127:374e83. [67]  Skuta R, Mat ejka V, Foniok K, Smy´kalov a A, Cvejn D, Gabor R, et al. On P-doping of graphitic carbon nitride with hexachlorotriphosphazene as a source of phosphorus. Appl Surf Sci 2021;552:149490. [68] Sun J, Gu X, Zhang S, Coquelle M, Bourbigot S, Duquesne S, et al. Improving the flame retardancy of polyamide 6 by incorporating hexachlorocyclotriphosphazene modified MWNT. Polym Adv Technol 2014;25:1099e107. [69] Osada Y, Hashidzume M, Tsuchida E, Bell AT. Polymerization of phosphazene crystal by plasma-exposure. Nature 1980;286:693e4. [70] Kuntz C, Kuhn C, Weickenmeier H, Tischer S, B€ ornhorst M, Deutschmann O. Kinetic modeling and simulation of hightemperature by-product formation from urea decomposition. Chem Eng Sci 2021;246:116876. [71] Li B, Peng W, Zhang J, Lian J-C, Huang T, Cheng N, et al. Highthroughput one-photon excitation pathway in 0D/3D heterojunctions for visible-light driven hydrogen evolution. Adv Funct Mater 2021;31:2100816. [72] Dong G, Ai Z, Zhang L. Efficient anoxic pollutant removal with oxygen functionalized graphitic carbon nitride under visible light. RSC Adv 2014;4:5553e60. [73] Pan H, Zhang H, Liu H, Chen L. Interstitial boron doping effects on the electronic and magnetic properties of graphitic carbon nitride materials. Solid State Commun 2015;203:35e40. [74] Han Q, Hu C, Zhao F, Zhang Z, Chen N, Qu L. One-step preparation of iodine-doped graphitic carbon nitride nanosheets as efficient photocatalysts for visible light water splitting. J Mater Chem 2015;3:4612e9. [75] ZhangY,MoriT,YeJ,AntoniettiM.Phosphorus-doped carbon nitride solid: enhanced electrical conductivity and photocurrent generation. J Am Chem Soc 2010;132:6294e5. journal of materials research and technology 2022;18:3319e33353334 [76] Li B, Si Y, Fang Q, Shi Y, Huang W-Q, Hu W, et al. Hierarchical self-assembly of well-defined louver-like P-doped carbon nitride nanowire arrays with highly efficient hydrogen evolution. Nano-Micro Lett 2020;12:52. [77] Thomas LC, Chittenden RA. Characteristic infra-red absorption frequencies of organophosphorus compoundsdV: phosph-orusdcarbon bonds. Spectrochim Acta 1965;21:1905e14. [78] Baumanis C, Bahnemann DW. TiO2 thin film electrodes: correlation between photocatalytic activity and electrochemical properties. J Phys Chem C 2008;112:19097e101. [79] Beranek R. (Photo)electrochemical methods for the determination of the band edge positions of TiO 2 -based nanomaterials. Adv. Phys Chem 2011;2011:786759. [80] Hankin A, Bedoya-Lora FE, Alexander JC, Regoutz A, Kelsall GH. Flat band potential determination: avoiding the pitfalls. J Mater Chem 2019;7:26162e76. journal of materials research and technology 2022;18:3319e3335 3335