Synthesis of graphene oxide and nitrogen-doped graphene oxide by nanosecond pulsed laser ablation of graphene in liquid for fiber optic gas sensing application

Heteroatom-doped graphene oxide has a wide range of applications in bio-imaging and sensing. In this work, Graphene Oxide (GO) and Nitrogen-doped GO (NG) were synthesized by laser ablation of Graphene in ethanol. The dopant Diethylenetriamine (DETA) is used in different amounts for different nitrogen concentrations. Optical, morphological, structural, and elemental composition studies were done by UV–vis spectroscopy, FT-IR, FE-SEM, XRD, Raman, and EDAX analysis, respectively. The nitrogen doping on the surface of GO was confirmed by FT-IR and EDAX studies. Upon laser ablation with fundamental wavelength, the graphene is converted to spherical GO nanoparticles, and nitrogen doping is done to produce porous nano coral structured NG nanoparticles. The sensitivity and selectivity of GO and NG for ammonia, ethanol, and acetone target gaseous were investigated and compared. NG sample shows excellent sensitivity and selectivity towards acetone gas. And the Nitrogen-doped graphene oxide can be considered an ideal material for gas-sensing applications.


Introduction
Gas sensors can monitor toxic gases to prevent their proliferation in the atmosphere permanently.Monitoring the concentration of toxic gases in the atmosphere, such as ammonia, ethanol, acetone, and methanol, is crucial for ensuring human safety in various sectors, such as agriculture and industry.Consequently, there is a pressing need to develop cost-effective, low-power, and highly sensitive sensors capable of detecting and measuring these hazardous gases in the environment [1,2].Even though there are various gas sensing techniques, fiber optic gas sensors have the advantages of room temperature sensing and their ability to detect gases at explosive and inaccessible locations [3].
Recently, there has been a growing interest in utilizing carbon-based nanomaterials, including carbon nanotubes and graphene, for gas detection applications at room temperature.These carbon nanomaterials exhibit excellent potential for electrochemical and gas sensing applications, primarily attributed to their remarkable characteristics, including high surface-to-volume ratio, enhanced carrier mobility, and superior gas adsorption capacity.While carbon nanotube gas sensors have been extensively investigated for detecting various gases, they suffer from limitations such as low sensitivity, extended response time, and poor reproducibility, which can be attributed to the assembly process and material purity [5][6][7].
In contrast, graphene-based materials offer significant advantages in developing high-performance gas sensors, particularly for detecting gases at lower operating temperatures.Graphene oxide (GO) and nitrogendoped graphene oxide (NG) have gained considerable attention as active materials for gas molecule detection.Incorporating oxygenated sheets and functional groups in graphene structures results in a significantly increased surface area, leading to enhanced gas detection capabilities compared to its non-oxygenated counterpart.This improvement can be attributed to the adsorption of gas molecules on the surface of the graphene sheets.Nitrogen doping enhances graphene's electron-donor properties and binding ability, offering intriguing possibilities in high-frequency semiconductor devices, catalysis for energy conversion and storage, and improved biocompatibility for biosensing applications [8][9][10][11].Moreover, NG's exceptional surface activity enables the adsorption of toxic gases.Tuning the structural and optical properties of these materials, the gas sensing capacity can be improved [12].
Several conventional chemical methods are used to synthesize GO and NG [11,13,14].Compared to alternative methods of fabricating nanostructures, Pulsed Laser Ablation in Liquids (PLAL) stands out for its straightforwardness, speed, and eco-friendliness, devoid of harmful chemicals.Its simple experimental setup, devoid of costly vacuum chambers or intricate procedures, adds to its appeal.Referred to by various names such as Pulsed Laser Induced Reactive Quenching (PLIRQ), Laser Ablation in Liquid Media (LALM), Laser Ablation Synthesis in Solution(LASiS), and Laser Pyrolysis-Pulsed Laser Fragmentation (LP-PLF) among researchers, PLAL has garnered significant attention, particularly in creating diverse nanostructures and graphene-based materials.A notable advantage lies in its versatility, allowing the combination of different materials and liquids to generate a wide array of nanostructures.Additionally, precise control over nanostructure characteristics like size, shape, distribution, and structure is achieved by adjusting laser parameters such as wavelength, pulse width, repetition rate, energy, and fluence.The duration of irradiation during or after ablation emerges as a crucial factor influencing nanostructure size [15].In terms of applications, nanostructures synthesized through the PLAL technique exhibit remarkable stability and are well-suited for utilization in biological and biochemical applications.Apart from these facts, laser-ablated materials can perform several post-ablation methods and reirradiation, after completion to make the material more suitable for several applications.The post-ablation leads form to water-soluble, nontoxic, and more functionalized materials, which will be suitable for Biological and optoelectronic applications.Moreover, these nanostructures can be created without capping ligands on their surfaces and offer the possibility of later-stage functionalization if needed [7,14,15].It presents several advantages over traditional synthesis methods, such as reduced reliance on chemical precursors and surfactants, faster synthesis times, and elimination of pre-and post-sintering processes and furnaces.PLAL takes very small time (maximum of an hour) to synthesize nanoparticles with high purity [18].
In this study, an innovative approach was adopted utilizing graphene submicron flakes dispersed in ethanol instead of employing a target or pellet.This method offers distinct advantages over target ablation techniques by covering a broader area in shorter intervals.Unlike target ablation, where only a small surface area is irradiated, the use of an unfocused laser beam (0.95 cm diameter) directed into the beaker facilitates extensive ablation across a larger area within a brief timeframe.Consequently, this approach enables the complete ablation of the prepared precursor solution and doping in a single step, a noteworthy efficiency.These attributes collectively position the PLAL method as highly recommended for synthesis, showcasing its capacity to achieve comprehensive ablation and doping in a more efficient manner compared to traditional target-based methods.
The use of ammonia and acetone as widely employed chemicals in the industry presents concerns regarding potential leakage and exposure to toxic gases, posing risks to both humans as well as the ecosystem.To address this issue, researchers have focused on fabricating sensors for the detection of toxic gases by modifying a small portion of a PMMA (polymethyl-methacrylate) plastic optical fiber's clad.The development of sensors based on the clad modification of PMMA plastic optical fibers presents a potential solution for detecting and monitoring toxic gases, mitigating risks associated with ammonia and acetone leakage [4,5,19].Optical fiber sensors operate on the principle of changing the refractive index of the modified clad, which in turn changes the output optical signal, indicating variations in the transmitted light intensity through the fiber [20].
In the course of this research, graphene oxide (GO) and nitrogen-doped graphene oxide (NGs) were synthesized separately via a single-step nanosecond pulsed laser ablation of graphene in a liquid medium.Notably, the synthesized GO and NGs exhibited distinct morphological and structural properties, with variations in size and shape.The selection of samples with superior structural characteristics paved the way for gas sensing applications.The exploration of gas sensing capabilities using readily prepared graphene-derived nanomaterials represents a relatively uncharted territory, offering a promising avenue for researchers.The present study uniquely contributes to this field by introducing a novel indigenous method for the preparation of these nanomaterials, resulting in exceptional gas-sensing properties.Our investigation focused on evaluating the gas-sensing potential of GO and NGs for detecting ammonia and acetone at various gas concentrations (ranging from 0 to 500 ppm) at room temperature.This research not only broadens our understanding of graphenebased sensors but also demonstrates the practical feasibility of using clad-modified fiber optic gas sensors for real-world applications, further emphasizing the novelty and significance of this work.

Experimental section
2.1.Synthesis of GO and NG An Nd:YAG nanosecond pulsed laser with a fundamental wavelength of 1064 nm ( pulse duration of 10 ns, repetition rate of 10 Hz, and beam diameter of 0.95 cm) is used (Spectra-Physics, Quanta-Ray, Model Pro 230-10) for pulsed laser ablation.Figure 1 depicts the experimental configuration.An unfocused laser beam with a pulse energy of 700 mJ was used to irradiate a 50 mL beaker containing 40 mL of a graphene precursor solution (0.8 mg mL −1 ) (sub-micron sized graphene nanoplatelets-Alfa Aesar; ethanol-spectrum) for 1 h.In a sealed chamber, a glass beaker with a lid was used to laser ablate a graphene solution placed nearly 90 cm from the laser source.A pyroelectric energy sensor (Coherent, Model: FieldMaxII-Top) measures the laser pulse's output energy.To achieve uniform ablation, magnetic stirring was used at a moderate speed of 480 rpm to prevent the gravitational settling of the graphene solution.
The graphene solution's color changes from black to a transparent, pale yellow after 60 min of laser exposure, indicating that all of the graphene has been ablated.The ethanol was removed by drying the ablated graphene solution (synthesized GO) at room temperature for three days.A brown material called GO was left behind at the bottom of the glass vessel after the graphene solution dried.Then, to separate the GO powder from its supernatant, 100 mL of double-distilled water was added; it was sonicated for 35 min and centrifuged at 15,000 rpm for 15 min.For the preparation of Nitrogen-doped graphene oxide(NG), the nitrogen dopant, DETA(Diethylenetriamine), was added dropwise during the ablation after 35 min (different amounts for different concentrations like 3 mL for NG3 and 5 mL for NG5).Then the ablation continues for 1 h.The color of the solution turned brownish yellow, and for powder sample preparation, the procedure followed that for GO (figure 2).The collected GO and NGs materials were dried at 70 °C overnight to remove the excess moisture content.Without any additional treatment, the obtained samples were sent for different characterizations.

Experimental setup for gas sensing
At room temperature, the synthesized NG was employed as modified clad materials in fibre optic gas sensing for acetone and ammonia gas.A 30 cm PMMA multimode plastic optical fiber was used.Without harming the fiber's core, the cladding portion at the center of the optical fiber was manually etched out to a length of about 3 cm.The etched area was examined under an optical microscope to check the clad removed area's surface uniformity.With a dip time of around two min, the sensing material NG5 was coated on the etched region [21].
The clad-removed portion was covered with the NG colloidal solution after it had dried and thickened.The optical fiber's thickness was measured using an optical microscope (Olympus brand, attached with the Microhardness Tester: Shimadzu, HMV-2) before and after coating.After drying at room temperature, the sensing region was put into the gas chamber of the fiber optic gas sensor setup for gas sensing.A fiber optic spectrometer (model: EPP-2000, StellarNet Inc., USA) with a white light source whose wavelength ranges from 100 to 2000 nm makes up the experimental setup for the gas sensor.The fiber terminals that connected the spectrometer and the light source to the sensor region were placed in separate gas chambers.The schematic diagram for the experimental setup for a fiber optic gas sensor is shown in figure 3. Using an HTC-1 hygrometer digital temperature and humidity meter, all gas sensing experiments were conducted at 25 °C room temperature and a relative humidity level of 55% [20].
The target gases, ethanol and ammonia were prepared by serial dilution method for the concentration 0 to 500 ppm for the gas sensing experiments.A stock solution containing 1000 ppm was made from the concentrated acetone and ammonia assay.Using the dilution factor equation, a series of required vapor solutions mixed with double distilled water in concentrations ranging from 0 to 500 ppm were created from this stock.

Characterization techniques
The Field Emission Scanning Electron Microscope ( Zeiss Sigma FESEM w/ EDX & EBSD) was used to characterize GO and NG's morphological and structural characteristics, and EDAX provides the elemental composition.By using X-Ray Diffraction (XRD, Brucker, Model: Ts55, Germany with Cu K radiation = 1.54056Å ) and Raman spectroscopy (WITec GmbH, Ulm, Germany Alpha300RA AFM & RAMAN, with 532 nm excitation wavelength), the structural and fundamental characteristics of GO and NGs were examined.Using KBr as a mulling agent and Fourier Transforms Infrared Spectroscopy (FT-IR, Thermofisher scientific) of GO and NGs confirmed the functional groups on and within the samples.A spectrophotometer (JASCO, V -670 ) was used to capture the optical absorption spectra.Using the spectrofluorometer (JASCO, FP-8500), the absorption spectra of the samples were obtained.
Clad-modified PMMA plastic optical fiber was used to sense ammonia and ethanol gas at various concentrations (0-500 ppm).The SpectraWiz software recorded the characteristic curve, and the corresponding data were plotted to examine the sensitivity and selectivity of the gas sensor.

The sensing mechanism of clad modified fiber optic gas sensor
Fibre optic sensors can detect toxic gases when a target gas is introduced by using a phenomenon known as the alteration in leaked light intensity through a modified cladding material.Interacting with the modified cladding region, the target gas causes a modification in the intensity of the guided signal along the core of the optical fiber.The change in intensity results in a corresponding change in the output spectrum intensity, which impacts the sensitivity and selectivity of the target gas sensor [22].
In the present work, the fiber optic sensor, the cladding medium, which is composed of synthesized nanoparticle (NG5), has a refractive index higher than that of the core material; this condition leads to partial reflection at the core/clad interface, and light enters into the modified cladding (leaky mode), the percentage of light reflected depends on the refractive index of core and cladding.
However, the refractive index of the gas in the sensing chamber is much smaller than that of the modified cladding.So, the ray undergoes total internal reflection, and the light will re-enter the core.The ray's intensity may change according to the change in the refractive index of the sensing material and the gas to be sensed [23].

FESEM analysis
Figures 4(a)-(f) shows the synthesized GO and NGs FESEM image and corresponding histogram.PLAL synthesized GO has a sphere-like morphology formed by the aggregation of the small spherical nanoparticle [24].In the size distribution, we can see that most particle lies under 100-200 nm.G K Yogesh et al already reported that two adjacent spheres of GO could fuse, and bigger spheres can be formed, which is why bigger nanospheres in the GO sample [17].The NG samples show nano coral structure with pores on the surface.Pores nature is more clearly visible in NG5 samples than in NG3 samples.All the samples possess a higher degree of agglomeration, and the higher degree of aggregation is increased ablation time and the time taken for room temperature evaporation.The varied diameter and agglomeration of nanoparticles may be due to the photothermal fragmentation of submicron-sized graphene in ethanol.Thus, it can be ensured that PLAL plays a significant role in forming GO and NG's.EDAX spectra of NGs confirm the nitrogen doping in GO.The EDAX spectra of GO have an atomic percentage of 93.60 % of carbon and 6.40% of oxygen NG3 sample consists of 7 % nitrogen, 77 %carbon and 14.77 % of oxygen and NG5 has an atomic percentage of 8.90 nitrogen and 75.64 % carbon and 15.46 % oxygen, which confirm the formation of NG samples.

XRD analysis
Figure 5 shows XRD patterns of precursor graphene(a), GO (b), and NGs (c).The XRD pattern of graphene exhibited peaks around 26.5°, 43.9°, 54.6°and 77.6°, which corresponds to the (002), (101), (004), and (110) planes, respectively, of the hexagonal graphitic structure.(JCDPS 89-7213, d(002) = 0.335 nm).After laser irradiation, a low-intensity peak at 2Ѳ = 14.1°was observed, corresponding to the (100) plane of GO with an interplanar spacing of 0.61 nm [16].This weak and broadened peak may be due to the addition of an oxygenous group, trapped moisture content, and the disruption of the graphitic structure by laser irradiation.By doping nitrogen, the peak intensity drastically decreases, and the GO tends to be amorphous.In the NG5 sample, the broadened peak corresponds to 2Ѳ ∼ 23.64 °, indicating the sample's poor crystallinity.The peak blue shift shifts as the nitrogen concentration increases [25].
The average crystallite size can be calculated from the Deby-Scherrer equation, Where D-average crystallite size k-shape factor with a value close to unity(0.9)λ-wavelength of the x-ray β-full-width half maxima (FWHM) in radians Ɵ-Bragg's angle in degrees The average crystalline size found to be varies from 34.4 nm (GO) to 13.57 nm (NG1 with interplanar distance, d = 0.63 nm) and 10.03 nm (NG2, d = 0.69 nm).

Raman spectra analysis
Figure 6 shows the Raman spectra of graphene and PLAL-synthesized GO (figure 6(b)) and NGs (figures 6(c) and (d)).Raman spectrum of graphene (figure 6(a)) shows the characteristic graphitic bands, D, G, and 2D bands.The breakdown of translational symmetry in the graphitic lattice or either defect may be responsible for the D band.The graphitic composition of the carbon material is confirmed by the G band, which arises due to the first order scattering of E 2g phonon from sp 2 carbon, and the 2D band is the overtone of the G band, which arises due to the double resonance of phonons.After laser ablation, the D band became prominent for the GO sample, the G band broadened, and it shifted to the higher wave number by 16 cm −1 [26].The 2D band of GO decreased in intensity, and the I 2D /I G ratio was 0.15, confirming the multi-layer nature of the sample.The I D /I G value increased around three times for the GO sample (0.5 to 1.3) because of the inclusion of oxygenious moieties by pulsed laser ablation, and it decreased the size of the sp 2 domain.The size of the sp 2 domain in the sp 3 lattice can be found by Tunistra-Koenig relation, For graphene, the size of the sp 2 domain is 38.45 nm, and for GO, it is 14.79 nm; the sp 2 domain size decreased by 2.5 times compared to the precursor graphene [27].The Raman spectra of the NG sample are shown in figures 6(b), (c); here, the G band shifts to a higher wavenumber than the precursor graphene.As doping concentration increases, a blue shift in the G band wavenumber is possibly due to the C-C and C-N bond length variation [11,28].The I 2D /I G ratio < 1 for NG3 (0.06) and NG5 (0.08) samples confirms the multi-layer nature of the prepared samples.Using equation (2), the size of the sp 2 domain in the sp 3 matrix is 24.03 nm for NG3 and 21.36 nm for NG5 which confirms the formation of GO and nitrogen-doped graphene oxide by pulsed laser ablation.The results confirm more sp 3 hybridized carbon than sp 2 hybridized carbon in GO and NGs compared to graphene [9].The peak positions of different bands are given in table 1.

FT-IR analysis
FT-IR spectra were recorded to identify various functional groups present in the samples.Figure 7(a) shows the FTIR spectrum of GO.It displayed a band at 3437 cm −1, representing the stretching vibration of the hydroxyl group.The 2850-3100 cm −1 absorption band corresponds to the asymmetric and symmetric stretching vibration of aliphatic carbon (C-H).The graphitic structure skeleton peak shows at 1627 cm −1 for sp 2 aromatic C=C stretching vibration.The absorption peak at 1720 cm −1 corresponds to C=O vibration (carboxylic functional group).An absorption peak at 808 cm −1 represents C=C bending vibration.Several researchers reported similar kind of vibration peak positions in synthesizing GO by chemical roots, and all these peaks confirmed the formation of GO from graphene by PLAL [16,27].
Figure 7(b) represents the FTIR spectrum of nitrogen-doped graphene oxide samples.The absorption peaks are almost similar to the GO sample, with some extra peaks corresponding to nitrogen functional groups.The absorption peak at 1388 cm −1 indicates the functional group for C-H bending vibration.A peak at 1040 cm −1 represents the epoxy functional group at the surface(O-C=O).The nitrogen-containing functional group is present at 1108 cm −1 , confirming the doping; along with this, the increment in the spectral intensity of O-H stretching at 3428 cm −1 also indicates the sample contains a nitrogen functional group in the material's surface [8,9,11].

UV-Vis analysis
Figures 8(a)-(c) shows the UV-Vis absorbance spectra of graphene oxide and nitrogen-doped graphene oxide samples.All peaks corresponding to different transitions lie within 200-400 nm, indicating no interaction of the formed GO/NGs with the laser radiation (1064 nm).Graphene oxide samples show peaks corresponding to ππ * and n-π * transitions of graphitic carbon around 224 nm and 246 nm, respectively.A shoulder peak around 300 nm is due to the n-π * transition of C=O of sp 3 hybridized carbon, which shows the formation of graphene oxide by PLAL [16].
Peak positions and shoulder peaks of NGs also have the same transition bands, with some peak shifts.NG3 shows the transition bands corresponding to π-π * and n-π * of C=C at 221 nm and 265 nm, respectively, and the transition band corresponding to C=O is at 309 nm.In the case of the NG5 sample, the peak of graphitic carbon π-π * transition shifted to 223 nm and 266 nm for its n-π * transition.The shift in the π-π * region of graphitic carbon may be due to the doping of nitrogen or maybe due to the reduction of graphene oxide [12].
The band gap of GO and NG samples are found out by Tauc-plot.The PLAL synthesized GO has a better bandgap than the reported values.The band gap of GO is obtained as 4.25 eV, and by doping, the band gap is increased by a higher concentration of nitrogen [29].The increase in bandgap(4.60eV for NG5) by increased nitrogen concentration may be due to the substitution effect of nitrogen (because of 8% of N doping.),or it can happen due to the combined effect of the sublattice symmetry breaking; the band gap is greatest when the dopants are positioned at the same sub lattice points of graphene [30].These interesting results indicate that bandgap tuning can be done for various opto-electronic applications.Thus, PLAL is a suitable synthesis method for such materials.
The prepared samples(GO and NGs) have higher bandgap compared to the reported results.It maybe due to the following reason.The band gap of graphene oxide is known to be variable, influenced by synthesis techniques, dopants, and structural alterations.Typically reported in the range of 3 to 4 eV according to research by V. Guptha et al, the specific value can be influenced by oxidation levels, present functional groups, and other chemical modifications.Higher degrees of oxidation, indicated by a higher C/O ratio, tend to result in a greater bandgap, as evidenced by Lundie et al's theoretical work citing a bandgap of 6.5 eV.Hsin-Cheng Hsu et al noted that increased oxidation disrupts the basal plane of graphene, introducing new oxygenated functional groups.Their findings showed a substantial band gap increase to 4.2 eV [31][32][33].
In this study, prolonged laser ablation using an unfocused 700mJ energy laser is anticipated to induce excessive graphene oxidation, resulting in a higher prevalence of oxygenated functional groups.This outcome is expected to elevate the C/O ratio similarly to the findings reported in the aforementioned studies.Similar studies in ns-PLAL a higher proportion of oxygen-containing functional groups is found on the basal plane compared to the edges of GO and NGs [16].Quantum confinement also play a role in increased bandgap due to the reduction of size by doping.The degree of oxidation and addition of oxygen containing functional groups in the structure are well evident from the FT-IR and Raman analysis.

Gas sensing
The gas sensing properties of synthesized nano-sized particles were studied for GO and NG5 for various concentrations of acetone and ammonia, like 100 ppm, 200 ppm, 300 ppm, 400 ppm, and 500 ppm, prepared by series dilution method [34].
The refractive index (n) of prepared GO and NG5 nanoparticles is found to be 2.17 and 2.13, respectively, which is calculated using the Mosse relation as follows.
Where E g is the bandgap of the prepared sample.
The refractive index of the modified clad (n = 2.17 for GO and 2.13 for NG) is much greater than the refractive index of the core (n = 1.492); there is a leaky mode.The light partially penetrates through the cladding region and partly reflects from the boundary between the core and cladding.However, compared to the modified cladding, the refractive index of the gas in the sensing chamber is much lower.As a result, the ray experiences total internal reflection and returns to the core.The ray's intensity may alter depending on how the gas is sensed, and the sensing material's refractive indexes change.
The characteristic spectra maxima were detected at 650 nm, typical of PMMA optical fiber.The intensity was reduced for different concentrations of acetone and ammonia for GO and NG samples.However, the photon count decrement was low for other higher wavelengths as more light may escape at higher wavelengths, so the intensity at higher wavelengths is almost constant, showing that the synthesized particle showed good spectral response variation to different concentrations of gases [21,34,35].Gas sensitivity can be defined as the change in spectral intensity with respect to different gas concentrations.Here, sensitivity was calculated for the maximum intensity peak at 650 nm for acetone, ethanol, and ammonia gaseous.The spectra of both GO and NG samples, target gases, show a monotonic decrease in intensity as concentration increases.
The sensitivity percentage of different gases can be calculated using the formula, Where, I o is the intensity at 0 ppm, and I g is the maximum intensity of the spectra in the presence of gas at different concentrations [36].
As the concentration of gas increases, a decrease in light reaches the spectrometer detector, which results in a decline in spectral response.Similar behaviour is shown when the clad is modified using the NG5 sample.
The spectral response of GO for ammonia is shown in figures 9(a), (b).The spectral intensity was high at 650 nm for all concentrations of ammonia gas. Figure 9(c) displays the estimated gas sensitivity by accounting for a plot between the intensity of the spectral peak (650 nm) and the concentrations of ammonia gas in the sample.The slope of the curve represents the sensitivity.The gas sensitivity of ammonia is found to be 35 counts/100 ppm, which is the slope of the spectral response graph.Similarly, by analyzing the spectral response of GO towards ethanol gas, as shown in figures 10(a), (b), GO shows 11 counts/100 ppm sensitivity.In the case of acetone gas (figures 11(a), (b)), the sensitivity is found to be 11 counts/100 ppm, and it is found that GO shows a better spectral response towards ammonia gas (figure 9(c)), which is almost 3 times the spectral sensitivity of acetone gas and ethanol gas(11 counts/100 ppm).These results are better compared to previous work done in GO [5].
Additionally, figure 12 shows a bar diagram representing the sensitivity percentage (%) for ammonia, ethanol, and acetone gases at each concentration for GO Sensitivity (%) is calculated by equation (4), and the  sensitivity percentage (%) values vary from 0.63 to 5.4 % for an increment of 100 ppm gas concentration for ammonia.The sensitivity percentage (%) for acetone gas varies from 0.62 to 2.18 %.Similarly, the percentage sensitivity of GO to ethanol is better than that of ammonia (2.9 %).This may be due to the hydrogen bonding between the functional group present in the material and the target gas.Sensitivity values are listed in table 2. From the linear fitted parameters of percentage sensitivity data, from table 2, the sensitivity percentage is increased with increase in gas concentration.For GO sample the slope of ammonia gas sensitivity percentage is  higher and in agreement with the sensitivity calculated by linear fit of intensity of spectral response to concentrations, GO is more selective to ammonia gas.It can be concluded that GO shows better sensitivity towards ammonia gas than acetone and ethanol.Figure 13 shows the selectivity of different gases for GO sample.From the sensitivity values GO is more sensitive towards ammonia gas.
Figure 14 shows the spectral response of the NG5 sample towards ammonia (figures 14(a), (b)).The spectral response of the ethanol target gas is shown in figures 15(a), (b), and the acetone gas is given in figures 16(a), (b).The spectral sensitivity of the ammonia, ethanol, and acetone gases for NG5 is found to be 26 counts/100 ppm, 23 counts/100 ppm, and 153 counts /100 ppm, respectively (figure 16(c)).For NG5, ammonia spectral sensitivity is almost six times higher than the other two gases.Furthermore, the spectral sensitivity (%) is found, and it is given as a bar diagram (figure 17), and percentage sensitivity values are given in table 3. The fitting parameters discussed in table 3. From table 3 the sensitivity percentage is increased with increase in gas concentration.For the sample NG the slope of the linear fit is higher for acetone gas and well agreement with counts/ ppm sensitivity, and NG is more selective to acetone.Here, the sensitivity percentage varies from 2.2 to 5.4 % for ammonia, 0.2 to 8.9 % for ethanol, and 6.2 to 26.2 % for acetone for the different concentrations.The bar diagram for selectivity (figure 18) of material towards different gases were drawn and the nitrogen doped samples were more selective to acetone gas.
Eight months later, the sensors for both GO and NG5 were retested across ammonia and acetone vapors (0-500 ppm).The sensitivity remained consistent, indicating stable performance in ambient conditions.Humidity changes also accounted in this experiment, which reveals that increased humidity levels can modify the sensor's response, particularly affecting the accuracy of detecting gases by altering the refractive index of GO and NG gas sensors.Elevated humidity tends to reduce sensitivity and selectivity in fiber optic gas sensors.This decline may be due to the hydrophilic nature of both GO and NG materials, limiting gas adsorption and subsequently reducing sensor output intensity.All fitting parameters for spectral response of GO and NG5 for target gases are given in supporting information (table S1).Comparing the spectral response towards different gaseous for GO and nitrogen-doped graphene oxide (NG), the nitrogen-doped sample has improved spectral response and sensitivity towards the toxic gaseous.Here, on comparing GO and NG5, both have similar sensitivity towards ammonia gas.But by doping with nitrogen, NG5's sensitivity value is around 10 times better than the sensitivity of GO for acetone gas, which suggests that the graphene oxide doped with heteroatoms is more significant for sensing toxic gases like acetone ethanol and ammonia.It can be explained in terms of the size of GO and NGs.GO's larger size may be the reason for reducing the gas sensing capacity for different toxic gaseous.Whereas the size reduced, heteroatom doped GO is highly sensible towards Acetone gas.
Coming to the morphological analysis, NGs have a porous structure, and this increases the surface reactivity and helps to increase the adsorption of the gas molecule on the surface, which alters refractive index of the modified clad and thus increases sensitivity for all the three gaseous studied.Functional groups play a pivotal role in the response of materials like graphene oxide (GO) and nitrogen-doped graphene (NG) to gases like ammonia and acetone.The introduction of nitrogen atoms into the graphene lattice of NG increases surface reactivity and creates defects and polar functional groups.These nitrogen-induced functional groups enhance sensitivity by forming hydrogen bonds with ammonia molecules, explaining the slight sensitivity increase compared to GO.When it comes to acetone, nitrogen-doped graphene exhibits superior sensitivity due to its heightened surface reactivity, increased density of polar functional groups from nitrogen doping, and specific interactions such as strong hydrogen bonding and Lewis Acid-Base interactions with acetone molecules.These factors collectively limit GO's sensitivity to acetone gas.So it can be concluded that pulsed laser ablated GO and NGs are promising candidates for fiber optic gas sensors, and the nitrogen-doped graphene shows better response and good sensitivity towards the toxic gaseous.The small size and porous nano coral structure of nitrogen-doped graphene oxide help to absorb the gas more efficiently, improved the sensing capabilities of the clad-modified optical fiber gas sensor.Table 4 shows literatures similar to fiber optic gas sensor with GO and its derivatives.From the table, the present work shows better gas sensitivity for GO and derivatives and NG material sensing is a new material in clad-modified fiber optic gas sensor [5].

Conclusion
GO and NGs were successfully synthesized by the nanosecond PLAL method.The surface morphology of the GO and NGs were determined using FESEM.The GO NPs have a spherical structure, and NGs with coral nanostructure are porous structure.EDAX and FT-IR confirm the nitrogen doping in the sample.XRD and Raman show the crystallinity decreases with doping.From the XRD data, the average crystalline size is found to be 10-34 nm with interplanar spacing 0.61 to 0.69 nm.The multi-layer nature and reduced size of the sp 2 hybridized carbon domain in sp 3 matrix are evident from the Raman analysis.The shift in characteristic peak positions of Raman spectra represents the samples' GO formation and nitrogen incorporation.UV-Vis spectroscopy confirms that GO and NG's formation and bandgap can be tuned by nitrogen doping and excessive oxidation results in wide bandgap of the samples.GO and NGs were subjected to clad-modified fiber optic gas sensing, and the spectral response towards the target gases was improved by nitrogen doping.NG5 samples show better spectral response, selectivity, and sensitivity for acetone gas (26.23 %) than ethanol(8.3%) and ammonia gas(5.43%), and for GO sample, ammonia(5.4%) is more sensitive than acetone(2.18%) and ethanol(2.9%) gas.The sensitivity of the NG5 sample for acetone gas is 153 counts/100 ppm, which is almost 15 times better than GO's sensitivity of other gases studied.Similarly, for ammonia gas, NG5 materials show slightly better sensitivity.The selectivity of the gas sensor were considered and GO shows better selectivity for ammonia gas and NG5 shows better selectivity for acetone gas.Consistent long-term performance of the sensor and fair repeatability assure that NG5 is very stable.These facts show that nitrogen-doped graphene oxide is a promising candidate for carbon material-based fiber optic gas sensing.

Figure 1 .
Figure 1.The Schematic diagram of the experimental setup for nanosecond pulsed laser ablation.

Figure 3 .
Figure 3. Schematic diagram of the experimental setup for gas sensing.

Figure 4 .
Figure 4. FESEM images, EDAX spectra and particle size distribution of GO and NG NPs.
where I D -Intensity of the D band, I G -Intensity of the G band and l-Wavelength of laser used in Raman spectroscopy analysis (532 nm)

Figure 8 .
Figure 8. Absorbance spectra of GO (a) and NG (b), (c) samples.The band gap of related materials is in the inset figure.

Figure 10 .
Figure 10.Spectral response of GO nanoparticle for ethanol target gas (a), zoomed view of spectral response (b), spectral peak intensity against different concentrations with the slope for ethanol gas (c).

Figure 9 .
Figure 9. Spectral response of GO nanoparticle for Ammonia target gas (a),zoomed view of spectral response (b), spectral peak intensity against different concentration with slope for ammonia gas (c).

Figure 12 .
Figure 12.Histogram of percentage sensitivity of GO for acetone, ethanol, and ammonia gas.

Figure 11 .
Figure 11.Spectral response of GO nanoparticle for acetone target gas (a), zoomed view of spectral response (b), spectral peak intensity against different concentrations with the slope for acetone gas (c).

Figure 13 .
Figure 13.Sensor selectivity for different target gas for GO.

Figure 14 .
Figure 14.Spectral response of NG nanoparticle for ammonia target gas (a),zoomed view of spectral response (b), spectral peak intensity against different comcentaion with slope for ammonia gas (c).

Figure 15 .
Figure 15.Spectral response of NG nanoparticle for ethanol target gas (a), zoomed view of spectral response (b), spectral peak intensity against different concentrations with the slope for ethanol gas (c).

Figure 16 .
Figure 16.Spectral response of NG nanoparticle for acetone target gas (a),zoomed view of spectral response (b), spectral peak intensity against different concentrations with slope for acetone gas (c).

Figure 18 .
Figure 18.Sensor selectivity for different target gas for NG.

Figure 17 .
Figure 17.Histogram of percentage sensitivity of NG for ammonia ethanol and acetone gas.

Table 1 .
Peak position in Raman spectra of GO and NG samples.

Table 2 .
Percentage sensitivity values of GO.

Table 4 .
Literatures similar to fiber optic gas sensor with GO and its derivatives.