nanomaterials Article

A Tunable Photoluminescent Composite of Cellulose Nanofibrils and CdS Quantum Dots Qinwen Wang, Aimin Tang *, Yuan Liu, Zhiqiang Fang and Shiyu Fu State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, China; [email protected] (Q.W.); [email protected] (Y.L.); [email protected] (Z.F.); [email protected] (S.F.) * Correspondence: [email protected]; Tel.: +86-20-8711-1225 Academic Editors: John H. T. Luong and Sandeep Kumar Vashist Received: 4 July 2016; Accepted: 31 August 2016; Published: 7 September 2016

Abstract: The preparation of fluorescent nanocomposite materials with tunable emission wavelengths by combining cellulose nanofibrils (CNFs) with inorganic nanoparticles is important for promoting CNFs applications. A CNF/CdS nanocomposite was prepared via in situ compositing at room temperature on oxidized CNFs with CdS quantum dots. By controlling the –COOH/Cd2+ ratio on the CNF, the feeding time of Na2 S and the ultrasonic maturing time, the size of the CdS quantum dots on the CNF surface could be adjusted so that to obtain the CNF/CdS nanocomposite material with different fluorescent colors. The results indicated that the CdS particles quantized were evenly distributed on the CNF. The maximum average size of the CdS nanoparticles glowed red under the excitation of UV light was 5.34 nm, which could be obtained with a –COOH/Cd2+ ratio of 1.0, a Na2 S feeding time of 20 min, and an ultrasonic maturing time of 60 min. A series of CNF/CdS nanocomposite materials were obtained with CdS nanoparticle sizes varying from 3.44 nm to 5.34 nm, the emission wavelength of which varied from 546 nm to 655 nm, and their fluorescence color changed from green to yellow to red. This is the first time the fluorescence-tunable effect of the CNF/CdS nanocomposite has been realized. Keywords: cellulose nanofibrils; CdS; quantum size effect; photoluminescence performance

1. Introduction CdS is a typical II–VI semiconductor and its particles exhibit good optical and photocatalytic properties, and tunable photoluminescence [1]. The different photoluminescence performances of CdS nanoparticles have different potential applications in fields such as biomarker [2], photoelectric devices [3], sensors [4], and ion detection [5]. Researchers have found that the optical properties of CdS nanoparticles can be adjusted by controlling their size, size distribution and surface coating. For example, Pattabi et al. [6] studied the preparation and stability of thiophenol-capped CdS nanoparticles. They found that the size of the CdS nanoparticles could be changed by adjusting the concentration of the thiophenol stabilizer. With an increase in stabilizer concentration, the obtained CdS particle sizes decreased. The optical absorption edge showed a blue shift, thereby realizing the tunable fluorescence. Chen et al. [7] prepared CdS/ZnS core/shell nanocrystals with a single-source precursor. The fluorescence characteristics of the CdS nanocrystals were adjusted by controlling the growth layers of the ZnS shell. It was found that the emission peak of the nanocrystals showed a gradual redshift with an increasing number of ZnS shell growth layers. However, CdS nanoparticles are easily aggregated due to their high surface free energy. Therefore, control of the particle size and stability is a key issue that needs to be solved. At present, inorganic-organic composite materials exhibit better mechanical, optical and electrical properties and thermal stability than those of single materials, and they have attracted significant Nanomaterials 2016, 6, 164; doi:10.3390/nano6090164

www.mdpi.com/journal/nanomaterials

Nanomaterials 2016, 6, 164

2 of 12

attention in many areas, including electrical equipment, solar devices, coatings, fuel cells, and sensors [8]. Petrochemical polymer products, including polyvinylpyrrolidone [9], poly(ethylene glycol)-block-poly(ethyleneimine) [10], and perfluorinated sulfonic acid [11], can be used for compositing with CdS particles. However, for environmental reasons, the existing petrochemical products are being gradually replaced by bio-based polymer materials. Silk protein fiber [12] and bacterial cellulose [1] have advantages in environmental protection and degradability, and have also been used for the preparation of CdS composites. Due to their advantages, such as renewability, degradability, large specific surface area, and a high number of active groups, plant-based cellulose nanofibrils (CNFs) have been widely used in many fields, including flexible electronic devices, cosmetics, and food additives. There have been numerous reports [13–16] that the surface-active groups of CNFs can adsorb cations such as Mn, Cd, Fe, and Zn. This characteristic is very advantageous for the control of compositing sites of inorganic nanoparticles, as well as the preparation of CNF/inorganic nanocomposites with uniform particle size distributions, which is realized through the adsorption of active groups and precursor ions by an in-situ compositing method. In a previous study, Li et al. [1] prepared a bacterial cellulose/CdS nanocomposite material. They believed that the hydroxyl groups in cellulose were reaction sites that could be used to control the distribution of CdS. In addition, the photoluminescence performance was also investigated. It was found that the ultraviolet absorption and emission peaks of the bacterial cellulose/CdS nanocomposite material showed a blue shift compared to the CdS bulk material. The plant-based CNFs prepared by TEMPO oxidization and mechanical treatment have carboxyl groups (–COOH). However, the questions of whether these carboxyl groups have the same functions as hydroxyl groups and how to prepare fluorescence tunable plant-based CNF/CdS nanocomposites have not yet been answered. In this study, CdS quantum dots were composited on CNF via in-situ compositing to prepare a CNF/CdS nanocomposite material. The CdS particle size in the CNF/CdS nanocomposite material was adjusted by varying the ratio of active groups (–COOH) on the CNF surface and Cd2+ , the feeding time of Na2 S and the ultrasonic maturing time. Consequently, the photoluminescence performances of the obtained material were adjustable. These results provide a basis for the preparation of emission wavelength tunable CNFs/CdS nanocomposite materials, which are applicable for many applications, including biomarkers, photoelectric devices, sensors, and ion detection. 2. Results and Discussion In the preparation process of CNF/CdS nanocomposite via in situ compositing, a number of factors may have influences on the CdS particles size and size distribution and, thus, the photoluminescence performance of the CNF/CdS nanocomposite may change. The involved factors include the adsorption of Cd2+ precursor on CNFs, the generation of CdS and its compositing on the CNFs. This paper mainly studies the following three aspects, the -COOH/Cd2+ ratios, the Na2 S feeding time and the ultrasonic maturing time. 2.1. Effect of –COOH/Cd2+ Ratio Firstly, the effects of –COOH/Cd2+ ratios on the morphology structure of the CNFs/CdS nanocomposite material were studied. The experiment was carried out on the following conditions: the feeding time of Na2 S was 20 min, the ultrasonic maturing time was 60 min, and the –COOH/Cd2+ ratios were 0.5, 1.0, and 2.0. The results are shown in Figure 1. From Figure 1, the particle size and distribution of CdS were obtained using Nano Measurer software (Microsoft Corporation, Redmond, WA, USA). The results are shown in Figure 2. Figure 1 presents the distribution change of CdS nanoparticles on the CNFs with the variation of the –COOH/Cd2+ ratio. The black points are CdS nanoparticles. It could be seen from the transmission electron microscope (TEM) image that the CdS nanoparticles were all uniformly dispersed onto the CNFs for three CNF/CdS nanocomposites prepared with different –COOH/Cd2+ ratios. This indicated that the CdS nanocomposite materials with CNFs as organic carriers could be successfully prepared

Nanomaterials 2016, 6, 164

3 of 12

by an in situ compositing method. The generation position of CdS was controlled by the position of Nanomaterials 6, 164 3 of 12 –COOH on 2016, the CNFs. Nanomaterials 2016, 6, 164

3 of 12

Figure 1.1. The images of of CNF/CdS nanocomposites nanocomposites with with different different –COOH/CdS –COOH/CdS ratios; Figure The TEM TEM images Figure 1. The TEM images of CNF/CdS CNF/CdS nanocomposites with different –COOH/CdS ratios; ratios; 2+ 2+ 2+ (a) –COOH/Cd –COOH/Cd 2+= 0.5; (b)(b) –COOH/Cd = 1.0; –COOH/Cd = 2.0 (the of Na 2S was 2+ and 2+ = feeding (a) = 0.5; –COOH/Cd 1.0;(c) and (c) –COOH/Cd 2.0 (thetime feeding time of 2+ = 1.0;=and 2+ = 2.0 (the (a) –COOH/Cd2+ = 0.5; (b) –COOH/Cd (c) –COOH/Cd feeding time of Na 2S was 20 min, and the ultrasonic maturing time was 60 min). Na S was 20 min, and the ultrasonic maturing time was 60 min). 2 20 min, and the ultrasonic maturing time was 60 min).

Figure 2. The size distributions of the CdS nanoparticles with different –COOH/CdS ratios; Figure 2. The 2+size distributions of 2+the CdS nanoparticles with different –COOH/CdS ratios; 2+ = 2.0. Figure 2. The size of the CdS with different –COOH/CdS ratios; (a) –COOH/Cd = 0.5;distributions (b) –COOH/Cd = 1.0; andnanoparticles (c) –COOH/Cd 2+ 2+ 2+ (a) –COOH/Cd = 0.5; (b) –COOH/Cd = 1.0; and (c) –COOH/Cd = 2.0. (a) –COOH/Cd2+ = 0.5; (b) –COOH/Cd2+ = 1.0; and (c) –COOH/Cd2+ = 2.0.

Figure 2 shows the size distribution of the CdS nanoparticles. When the –COOH/Cd2+ ratio was Figure 2 shows the size distribution of the CdS nanoparticles. When the –COOH/Cd2+ ratio was 2+ ratio 0.5, the average sizethe of CdS nanoparticlesofin CNF/CdS nanocomposite was 4.27 nm, and the Figure 2 shows size distribution thethe CdS nanoparticles. When the –COOH/Cd 0.5, the average size of CdS nanoparticles in the CNF/CdS nanocomposite was 4.27 nm, and the particle distributed in thenanoparticles range of 2–9 nm. The particles with sizes of 3.3–4.7 was 0.5, size the was average size of CdS in the CNF/CdS nanocomposite wasnm 4.27accounted nm, and particle size was distributed in the range of 2–9 nm. The particles with sizes of 3.3–4.7 nm accounted for 50% of allsize CdS particles, indicating wideofsize In other with words, theof uniformity of the particle was distributed in thethe range 2–9distribution. nm. The particles sizes 3.3–4.7 nm for 50% of all CdS particles, indicating the wide size distribution. In other words, the uniformity of 2+ CdS particle was When the –COOH/Cd was 1.0, average size CdS was clearly accounted forsize 50% ofpoor. all CdS particles, indicating ratio the wide sizethe distribution. Inofother words, the CdS particle size was poor. When the –COOH/Cd2+ ratio was 1.0, 2+ the average size of CdS was clearly increased to 5.34 nm. Moreover, the particle sizes were fully distributed in the range of 4.0–7.5 nm. uniformity of CdS particle size was poor. When the –COOH/Cd ratio was 1.0, the average size of increased to 5.34 nm. Moreover, the particle sizes were fully distributed in the range of 4.0–7.5 nm. The distribution width wastolower thanMoreover, that of thethe sample with a –COOH/Cd ratio of 0.5. CdS was clearly increased 5.34 nm. particle sizes were fully 2+ distributed inTherefore, the range The distribution width was lower than that of the sample with a –COOH/Cd2+ ratio of2+0.5. Therefore, 2+ ratio the4.0–7.5 uniformity of CdS particle size distribution was improved. When the with –COOH/Cd ratio was 2.0, of nm. The distribution width was lower than that of the sample a –COOH/Cd the uniformity of CdS particle size distribution was improved. When the –COOH/Cd2+ ratio was 2.0, the average particle size of CdS was reduced to 4.01 nm. In this case, the particles with sizes of of 0.5. Therefore, the uniformity of CdS particle size distribution was improved. When the the average 2+ particle size of CdS was reduced to 4.01 nm. In this case, the particles with sizes of 2.3–4.3 nm accounted for2.0, 72%the of average particles,particle indicating further in the of –COOH/Cd ratio was size the of CdS wasimprovement reduced to 4.01 nm.uniformity In this case, 2.3–4.3 nm accounted for 72% of particles, indicating the further improvement in the uniformity of particle sizewith distribution. Dai etnm al. accounted [17] foundfor that theofsize of CdSindicating nanoparticles also increases with the particles sizes of 2.3–4.3 72% particles, the further improvement particle size distribution. Dai et al. [17] found that the size of CdS nanoparticles also increases with increasing precursor (CdClsize 2 and Na2S) concentration. In this study, the ratio of –COOH also and in the uniformity of particle distribution. Dai et al. [17] found that varying the size of CdS nanoparticles increasing precursor (CdCl2 and Na2S) concentration. In2+this study, varying the ratio of –COOH and 2+ Cd 2+ was with used increasing to control precursor the adsorption of2 S) Cdconcentration. onto the CNFs. As study, a result, the size the increases (CdClamount In this varying theofratio 2 and Na Cd was used to control the adsorption amount of Cd2+ onto the CNFs. As a result, the size of the 2+ 2+ generated CdS nanoparticles was adjusted. Moreover, the distribution and distribution uniformity of of –COOH and Cd was used to control the adsorption amount of Cd onto the CNFs. As a result, generated CdS nanoparticles was adjusted. Moreover, the distribution and distribution uniformity of 2+ CdSofnanoparticles thenanoparticles CNFs were was controlled through the localized adsorption of Cd 2+ by the size the generatedon CdS adjusted. Moreover, the distribution and distribution the CdS nanoparticles on the CNFs were controlled through the localized adsorption of Cd by 2+ ratio was 0.5, the amount of Cd2+ adsorbed by the CNFs was small. –COOH. When –COOH/Cd uniformity of thethe CdS nanoparticles on the CNFs were controlled 2+ through the localized adsorption of 2+ ratio –COOH. When the –COOH/Cd was 0.5, the amount of Cd adsorbed by the CNFs was small. 2+ 2+ Therefore, the average size of CdS nanoparticles in the CNF/CdS was The Cd by –COOH. When the –COOH/Cd ratio was 0.5, the amountnanocomposite of Cd2+ adsorbed bysmall. the CNFs Therefore, the average size of CdS nanoparticles in the CNF/CdS 2+nanocomposite was small. The compositing amount was relatively the –COOH/Cd ratio was nanocomposite increased to 1.0,was the was small. Therefore, the also average size oflow. CdSWhen nanoparticles in the CNF/CdS compositing amount was also relatively low. When the –COOH/Cd2+ ratio2+was increased to 1.0, the 2+ ratio was contentThe of compositing –COOH increased, to relatively an increase inWhen the amount of Cd 2+ adsorbed by the CNFs. small. amountleading was also low. the –COOH/Cd increased content of –COOH increased, leading to an increase in the amount of Cd adsorbed by the CNFs. After Na2S,ofthe number of generated CdStonanoparticles Meanwhile, the particle to 1.0, adding the content –COOH increased, leading an increase inincreased. the amount of Cd2+ adsorbed by After adding Na2S, the number of generated CdS nanoparticles increased. Meanwhile, the particle 2+ size also increased. When the –COOH/Cd 2+ ratio was 2.0, the content of –COOH on the CNFs further size also increased. When the –COOH/Cd ratio was 2.0, the content of –COOH on the CNFs further increased; thus, the numbers of composited CdS nanoparticles increased. In addition, due to the increased; thus, the numbers of composited CdS nanoparticles increased. In addition, due to the chemical anchoring effect of –COOH on CNFs, the adsorbed Cd2+ was uniformly distributed onto the chemical anchoring effect of –COOH on CNFs, the adsorbed Cd2+ was uniformly distributed onto the CNFs. Therefore, the number of generated CdS crystal nuclei increased. The particle size of CdS CNFs. Therefore, the number of generated CdS crystal nuclei increased. The particle size of CdS decreased while the distribution uniformity improved.

Nanomaterials 2016, 6, 164

4 of 12

the CNFs. After adding Na2 S, the number of generated CdS nanoparticles increased. Meanwhile, the particle size also increased. When the –COOH/Cd2+ ratio was 2.0, the content of –COOH on the CNFs further increased; thus, the numbers of composited CdS nanoparticles increased. In addition, due to the chemical anchoring effect of –COOH on CNFs, the adsorbed Cd2+ was uniformly distributed onto the CNFs. Therefore, the number of generated CdS crystal nuclei increased. The particle size of CdS decreased while the distribution uniformity improved. 2+ ratios Furthermore, the PL spectra of the CNF/CdS nanocomposites with different –COOH/Cd Nanomaterials 2016, 6, 164 4 of 12 were investigated. As shown in Figure 3, at an excitation wavelength of 365 nm, a strong emission 2+ ratios Furthermore, theof PL420,000 spectra of the was CNF/CdS nanocomposites –COOH/Cd peak with an intensity a.u. observed at 655 nmwith for different the CNF/CdS nanocomposite 2+ werewhen investigated. As shown in Figure 3, at1.0. an excitation wavelength of 365 nm, a to strong emission material the –COOH/Cd ratio was The emission peak blueshifted 613 nm when the 2+ peak with an intensity of 420,000 a.u. was observed at 655 nm for the CNF/CdS nanocomposite –COOH/Cd ratio was 2.0. However, the fluorescence intensity was at a maximum (1,020,000 a.u) material when the –COOH/Cd2+ ratio was 1.0. The emission peak blueshifted to 613 nm when the – at this time. When the –COOH/Cd2+ ratio was 0.5, the CNF/CdS nanocomposite had a very wide COOH/Cd2+ ratio was 2.0. However, the fluorescence intensity was at a maximum (1,020,000 a.u) at emission bandwidth at 500–625 nm, and the intensity was only 20,000 a.u. By comparing the data this time. When the –COOH/Cd2+ ratio was 0.5, the CNF/CdS nanocomposite had a very wide in Figure 2, it could be found that the CdS particle size distribution range was wide, the size and emission bandwidth at 500–625 nm, and the intensity was only 20,000 a.u. By comparing the data in 2+ ratio was 0.5. As a result, the number of CdS particles wasparticle small size when the –COOH/Cd Figure composited 2, it could be found that the CdS distribution range was wide, the size and number 2+ ratio was resulting CNF/CdS nanocomposite exhibited a wide fluorescence emission with low intensity. of composited CdS particles was small when the –COOH/Cd 0.5. As aband result, the resulting 2+ ratio increased, the size distribution uniformity of the CdS particles was When the –COOH/Cd CNF/CdS nanocomposite exhibited a wide fluorescence emission band with low intensity. When the 2+ ratio increased, the size distribution uniformity of the CdS particles was gradually –COOH/Cd gradually improved. Meanwhile, the number of composited CdS nanoparticles increased. Therefore, improved. Meanwhile, the number composited CdS nanoparticles increased. Therefore, the fluorescence emission band of the of CNF/CdS nanocomposite was gradually narrowed the and the fluorescence emission band of the CNF/CdS nanocomposite was gradually narrowed and the intensity was enhanced significantly. In addition, it has been reported [18] that a redshift phenomenon intensity was enhanced significantly. In addition, it has been reported [18] that a redshift occurs with increasing CdS particle size, as a result of the quantum size effect. When the –COOH/Cd2+ phenomenon occurs with increasing CdS particle size, as a result of the quantum size effect. When ratio was decreased2+ from 2.0 to 1.0, the average size of the CdS nanoparticles composited onto the the –COOH/Cd ratio was decreased from 2.0 to 1.0, the average size of the CdS nanoparticles CNFscomposited increasedonto fromthe 4.01 nmincreased to 5.34 nm, that the of the showed the redshift CNFs fromso4.01 nm to PL 5.34spectra nm, so that thematerial PL spectra of the material phenomenon. Therefore, this study, the PL spectra of the CNF/CdS consistent showed the redshift in phenomenon. Therefore, in this study, the PLnanocomposite spectra of thewere CNF/CdS with nanocomposite the size distribution results of CdS nanoparticles. were consistent with the size distribution results of CdS nanoparticles. 0.5 1.0 2.0 1020000

intensity (a.u.)

820000

620000

420000

220000

20000 400

450

500

550

600

650

700

wavelengh (nm) Figure 3. The PL spectra of the CNF/CdS nanocomposites with different –COOH/Cd2+ ratios

Figure 3. The PL spectra of the CNF/CdS nanocomposites with different –COOH/Cd2+ ratios (excitation wavelength 365 nm). (excitation wavelength 365 nm).

2.2. Effect of Na2S Feeding Time

2.2. Effect of Na2 S Feeding Time

In the condition of that the CNFs carboxyl contents was 1.63 mmol/g, –COOH/Cd2+ ratio was 2+ ratio controlled at 2.0 and maturing time was fixed atwas 60 min, effect of –COOH/Cd Na2S feeding time on was In the condition of ultrasonic that the CNFs carboxyl contents 1.63the mmol/g, the PL at spectrum the CNF/CdS nanocomposite material was Theof results showntime in on controlled 2.0 andofultrasonic maturing time was fixed at 60investigated. min, the effect Na2 Sare feeding Figure 4. the PL spectrum of the CNF/CdS nanocomposite material was investigated. The results are shown in

Figure 4.

Nanomaterials 2016, 6, 164 Nanomaterials 2016, 6, 164

5 of 12 5 of 12

40 min 30 min 20 min 10 min

1200000

1000000

intensity (a.u.)

800000

600000

400000

200000

0 400

450

500

550

600

650

700

wavelengh (nm) Figure 4. The PL spectra of the CNF/CdS nanocomposites with different Na2S feeding time (excitation Figure 4. The PL spectra of the CNF/CdS nanocomposites with different Na2 S feeding time (excitation wavelength 365 nm). wavelength 365 nm).

It could be seen from Figure 4 that the PL peak of the CNF/CdS nanocomposite was 605 nm (excitation nm) when time NaCNF/CdS 2S solution nanocomposite was 10 min. Moreover, thenm It could wavelength, be seen from365 Figure 4 thatthe thefeeding PL peak of of the was 605 emission wavelength, band appeared in nm) the range 500–650 nm time with aofserious phenomenon. When (excitation 365 whenofthe feeding Na2 S broadening solution was 10 min. Moreover, feeding time Na2S solution was prolonged to 20 min 30 amin, the PL peaks showed redshifts thethe emission bandofappeared in the range of 500–650 nmand with serious broadening phenomenon. at 613 and 640 nm,ofrespectively. If the feeding time oftoNa solution was further prolonged to 40 When thenm feeding time Na2 S solution was prolonged 202Smin and 30 min, the PL peaks showed min, the PL peak at 625 nm showed a blue shift. It was generally accepted that the high energy redshifts at 613 nm and 640 nm, respectively. If the feeding time of Na2 S solution was further prolonged luminescence (300–500 nm)showed of CdS was dueshift. to theItband while the low energy luminescence to 40 min, the PLband peak at 625 nm a blue was gap, generally accepted that the high energy band (500–700 nm) was related to the surface defects. The wavelengths of the PL peaks of the luminescence band (300–500 nm) of CdS was due to the band gap, while the low energy luminescence CNF/CdS nanocomposite in this study were all higher than 600 nm, indicating that they were mainly band (500–700 nm) was related to the surface defects. The wavelengths of the PL peaks of the CNF/CdS due to the surface defects emission. nanocomposite in this study were all higher than 600 nm, indicating that they were mainly due to the In order to reveal the reasons why the CNF/CdS nanocomposites prepared with different Na2S surface defects emission. feeding time shows different luminescence, TEM analysis and Nano Measurer software was used to In order to reveal the reasons why the CNF/CdS nanocomposites prepared with different Na S statistically calculate the CdS particle sizes. Figure 5 presents the size distribution of CdS 2 feeding time shows different luminescence, TEM analysis and Nano Measurer software was used to nanoparticles, which was obtained using the Nano Measurer software based on a TEM image of the statistically the CdS Itparticle sizes. Figure 5 presents sizein distribution of CdS nanoparticles, CNF/CdScalculate nanocomposite. indicated that the CdS particlethe sizes the CNF/CdS nanocomposite which was obtained using the Nano Measurer software based on a TEM image of the CNF/CdS were 3.76 nm and 4.01 nm when the feeding time of Na2S was 10 min and 20 min, respectively. The nanocomposite. indicated that the CdSincreased particle sizes in nm the when CNF/CdS nanocomposite were 3.76 nm average size ofItthe CdS nanoparticles to 4.86 the feeding time of Na 2S was 30 and 4.01With nm the when the feeding time 10 min 20min, min,the respectively. size 2S min. further prolonging ofof theNa Na 2Swas feeding timeand to 40 CdS particleThe sizeaverage decreased of the CdS nanoparticles increased to 4.86 nm when the feeding time of Na S was 30 min. With to 4.59 nm. This could be explained by the fact that when the feeding time of2 Na2S was less than 20the further prolonging ofdropping the Na2 S speed), feedingthe time to 40 min, the CdSCdS particle size decreased to 4.59 nm. This min (relatively fast number of generated crystal nuclei was large. However, could be explained by were the fact the feeding timetime, of Na S was less than 20 min (relatively the crystalline grains notthat fullywhen grown within a short so2the particle size was small. When the feeding time was prolonged to 30 min, the formation reaction of CdS nanoparticles was basically fast dropping speed), the number of generated CdS crystal nuclei was large. However, the crystalline complete, the average of the obtained CdS was at a maximum. If the feeding grains were and not fully grownsize within a short time, sonanoparticles the particle size was small. When the feeding time of Na 2S was further (relatively slow dropping speed), there waswas not enough S2−complete, during time was prolonged to 30 increased min, the formation reaction of CdS nanoparticles basically thethe formation of CdS. As a result, the average size of at theaCdS nanoparticles The of and averagereaction size of the obtained CdS nanoparticles was maximum. If thedecreased. feeding time 2−:Cd2+ molar ratios affected the growth 2 − studies of Martinez-Castanon et al. [19] showed the S and size Na2 S was further increased (relatively slow dropping speed), there was not enough S during the 2− concentration in precursors during of particles. It wasof observed the particle size increased the Snanoparticles formation reaction CdS. Asthat a result, the average size of with the CdS decreased. The studies 2− 2+ molar ratios in unit time. the reaction. The different feeding time of Na 2S made the different S :Cd 2 − 2+ of Martinez-Castanon et al. [19] showed the S :Cd molar ratios affected the growth and size of

particles. It was observed that the particle size increased with the S2 − concentration in precursors during the reaction. The different feeding time of Na2 S made the different S2 − :Cd2+ molar ratios in unit time.

Nanomaterials 2016, 6, 164

6 of 12

Nanomaterials 2016, 6, 164

6 of 12

(1)

(2)

Figure 5. (1) The size distributions of the CdS nanoparticles with different Na2S feeding time; and (2) Figure 5. (1) The size distributions of the CdS nanoparticles with different Na2 S feeding time; the TEM images of CNF/CdS nanocomposites with different Na2S feeding time; (a) 10 min; (b) 20 min; and (2) the TEM images of CNF/CdS nanocomposites with different Na2 S feeding time; (a) 10 min; (c) 30 min; and (d) 40 min (–COOH/Cd2+ ratio 2.0, ultrasonic maturing time 60 min). (b) 20 min; (c) 30 min; and (d) 40 min (–COOH/Cd2+ ratio 2.0, ultrasonic maturing time 60 min).

The work of Pattabi et al. [20] indicated that the absorption spectrum of CdS nanoparticles The work of Pattabi et al.particle [20] indicated that the absorption spectrumsize of CdS nanoparticles showed showed a blueshift when size decreased. As the particle decreased, the blueshift aincreased. blueshift This when particle size decreased. As the particle size decreased, the blueshift increased. was associated with the quantum confinement effect. According to [21], the bandThis gap was associated with the quantum confinement effect. According to [21], the band gap of a CdS quantum of a CdS quantum dot is in accordance with the Brus equation (Equation (1)) of the absorption band dot in accordance with the Brus equation (Equation (1)) of the absorption band edge movement: edgeismovement: 2 2

2

1.8e 2h π (1) 𝐸 = 𝐸g +h π∗2 2 − 1.8e2− 0.25𝐸Ry , 2𝑚 𝑅 − ε𝑅 E1 =1 Eg + − 0.25ERy , (1) ∗ R2 εR 2m −1 1 1 where 𝑚∗ = ( + )−1, where E1 is the band gap energy of the quantum dot, R is the radius of the 𝑚𝑒 𝑚ℎ 1 where m∗ dot, = mm + m1h mh are , where E1 is themass band of gap the quantum dot, R is the respectively, radius of the e e and quantum the effective anenergy excitedofelectron and excited hole, quantum dot, m mh are the bulk effective mass of an excited andrelated excitedtohole, respectively, Eg e and Eg is the band gap energy of the semiconductor, ERy is aelectron parameter the exciton binding is the band gap energy of the bulk and semiconductor, Ry is a parameter related to the exciton binding energy of the bulk semiconductor, h is Planck’sEconstant. energy of the bulk semiconductor, and h is Planck’s constant. According to the Brus equation, when the radius of quantum dot (R) is small enough, the band According the Brus equation, when the size, radius of quantum (R)shift is small enough, the band gap will increasetowith decreasing nanoparticle resulting in thedot blue of absorption peaks. It gap will increase with decreasing nanoparticle size, resulting in the blue shift of absorption peaks. could be seen from Figures 4 and 5 that the average size of the CdS nanoparticles was 4.86 nm when It be seen 4 and 5 that the average size of the CdS nanoparticles was 4.86 nm thecould feeding time from of NaFigures 2S was 30 min, and the PL peak of the CNF/CdS nanocomposite was 640 nm. when the feeding time of Na the PL of the nanocomposite When the feeding time of Na22SS was was 30 40 min, min, and 20 min, andpeak 10 min, theCNF/CdS average sizes of the CdS was 640 nm. When the feeding time of Na S was 40 min, 20 min, and 10 min, the average sizes 2 nanoparticles decreased to 4.59 nm, 4.01 nm and 3.76 nm, respectively. Accordingly, the PL peaks of of the CdS nanoparticles decreased to 4.59 nm, 4.01 nm and 3.76 nm, respectively. Accordingly, the PL the CNF/CdS nanocomposite showed blue shifts at 625 nm, 613 nm and 605 nm, respectively. peaks of thethe CNF/CdS showed blue shiftswere at 625innm, nmof and 605 nm, respectively. Therefore, PL peaksnanocomposite of the CNF/CdS nanocomposite the613 range 640–605 nm, with only Therefore, the PL peaks of the CNF/CdS nanocomposite were in the range of 640–605 nm, with only a a small difference occurring when the feeding time of Na2S was changed. These results indicated that small difference occurring when the feeding time of Na S was changed. These results indicated that the 2 varying the feeding time of Na2S in order to the size of the CdS nanoparticles could be changed by size of the thePL CdS nanoparticles could be changed by varying the feeding time of Na order tolarge. adjust 2 S in was adjust spectrum of the CNF/CdS nanocomposite. However, the adjustable degree not the PL spectrum of the CNF/CdS nanocomposite. However, the adjustable degree was not large.

2.3. Effect of Ultrasonic Maturing Time 2.3. Effect of Ultrasonic Maturing Time For these experiments, the –COOH/Cd2+ ratio was 1.0 and the feeding time of Na2S was 20 min. For these experiments, the –COOH/Cd2+ ratio was 1.0 and the feeding time of Na2 S The effect of ultrasonic maturing time on the PL spectrum of the CNF/CdS nanocomposite was was 20 min. The effect of ultrasonic maturing time on the PL spectrum of the CNF/CdS nanocomposite investigated. The maturing time in the ultrasonic setup was 40 min, 60 min, 80 min, and 100 min, was investigated. The maturing time in the ultrasonic setup was 40 min, 60 min, 80 min, and 100 min, respectively. The results are shown in Figures 6 and 7. respectively. The results are shown in Figures 6 and 7.

Nanomaterials 2016, 6, 164 Nanomaterials 2016, 6, 164

7 of 12 7 of 12

Figure 6. The fluorescence images of the CNF/CdS nanocomposites under different maturing time; (a) 40 min; (b) 60 min; (c) 80 min; and (d) 100 min.

Figure 6 illustrates the fluorescence image of the CNF/CdS nanocomposite obtained under different ultrasonic maturing time conditions (λ = 365 nm UV light). It could be seen that the CNF/CdS nanocomposite showed a red color under the excitation of the UV light only when the Figure 6. fluorescence images images of of the the CNF/CdS CNF/CdS nanocomposites under maturing time; Figuretime 6. The The fluorescence nanocomposites under different different maturing time; maturing was 60 min, while the composite showed a yellow-green color when the ultrasonic (a) 40 min; (b) 60 min; (c) 80 min; and (d) 100 100 min. min. (a) 40 min; (b) 60 min; (c) 80 min; and (d) maturing time was 40 min, 80 min, or 100 min. Figure 6 illustrates the fluorescence image of the CNF/CdS nanocomposite 40 min obtained under different ultrasonic maturing time conditions (λ = 365 nm UV light). It could 60 minbe seen that the 600000 min only when the CNF/CdS nanocomposite showed a red color under the excitation of the UV80light 100 min maturing time was500000 60 min, while the composite showed a yellow-green color when the ultrasonic maturing time was 40 min, 80 min, or 100 min. intensity (a.u.)

400000

40 min 60 min 80 min 100 min

600000 300000 500000 200000

intensity (a.u.)

400000 100000 300000 0 200000400

450

500

550

600

650

700

wavelengh (nm) 100000

Figure 7. The PL spectra of the CNF/CdS nanocomposites with different maturing time (excitation Figure 7. The PL spectra of the CNF/CdS nanocomposites with different maturing time (excitation wavelength 365 nm).0 wavelength 365 nm).

Figure 7 shows the400 PL spectra maturing times. 450 of the 500CNF/CdS 550 nanocomposite 600 650with different 700 Figure 6 illustrates theused fluorescence of the CNF/CdS nanocomposite obtained under The excitation wavelength was 365 image nm. It could be seen that the PL peak of the CNF/CdS wavelengh (nm) different ultrasonic time conditions (λ = 365 nmwas UV 40 light). could the be seen that the CNF/CdS nanocomposite wasmaturing at 560 nm when the maturing time min.ItWhen maturing time was 60 Figure 7. The PL spectra ofcolor the CNF/CdS nanocomposites withUV different maturing time (excitation nanocomposite showed a red under the excitation of the light only when the maturing time min, a strong emission peak at 655 nm could be observed in the PL spectrum. When the maturing wavelength 365 the nm).composite showed a yellow-green color when the ultrasonic maturing time was 60 min, while time was further prolonged to 80 min and 100 min, the emission peaks showed blueshifts at 580 nm was 546 40 min, min, or 100The min. and nm, 80 respectively. wavelength at 655 nm was red light, which is consistent with the result Figure 77 shows the PL spectra spectra of of the the CNF/CdS nanocomposite with maturing times. Figure shows the PL with different different of the fluorescence image with the UV lightCNF/CdS (Figure 6).nanocomposite When the maturing time was maturing 40 min, 80times. min, The excitation wavelength used was 365 nm. It could be seen that the PL peak of the CNF/CdS CNF/CdS The100 excitation wavelength was 365 nm.inItthe could seen thatshowed the PL peak of the or min, however, theused emission peak PLbespectrum an obvious blueshift nanocomposite was atat560 when thethe maturing timetime was was 40 min. When When the maturing time was 60 nanocomposite was 560nm nm when 40 min. the maturing time phenomenon. The emission wavelength wasmaturing in the range of 546–580 nm, which belonged to yellowmin, a strong emission peak at 655 nm could be observed in the PL spectrum. When the maturing was 60light. min,This a strong emission peakwith at 655 could be observed in the PL spectrum. When the green was also consistent thenm results of the fluorescence imaging. time was further prolonged to 80 min and 100 min, the emission peaks showed blueshifts at 580 nm maturing time was further prolonged to 80 min and min, the showed The morphology of CNF/CdS nanocomposite was100 observed byemission TEM. Thepeaks size of the CdSblueshifts particles and 546 nm, respectively. The wavelength at 655 nm was red light, which is consistent with the result at 580 nm and 546 nm, respectively. wavelength at 655 nmThe was red light, whichinisFigure consistent on the CNFs was calculated using the The Nano Measurer software. results are shown 8. It of thethe fluorescence image with the UV light (Figure When maturing time was min, 80 min, with resultthat of the the6). UV lightthe (Figure 6). When the 40 maturing time could be seen the fluorescence average sizesimage of the with CdS nanoparticles in the CNF/CdS nanocomposite were or min,80however, themin, emission peak the PLpeak spectrum showed an showed obviousanblueshift was100 40 min, min, or 100 however, the in emission in the PL spectrum obvious phenomenon. The emission wavelength was in the range of 546–580 nm, which belonged to yellowblueshift phenomenon. The emission wavelength was in the range of 546–580 nm, which belonged to green light. This was also consistent with the results of the fluorescence imaging. yellow-green light. This was also consistent with the results of the fluorescence imaging. The morphology of CNF/CdS nanocomposite was observed by TEM. The size of the CdS particles on the CNFs was calculated using the Nano Measurer software. The results are shown in Figure 8. It could be seen that the average sizes of the CdS nanoparticles in the CNF/CdS nanocomposite were

Nanomaterials 2016, 6, 164

8 of 12

The morphology of CNF/CdS nanocomposite was observed by TEM. The size of the CdS particles 8 of 12 on the CNFs was calculated using the Nano Measurer software. The results are shown in Figure 8. It could of the the CdSmaturing nanoparticles in the40CNF/CdS nanocomposite 3.68 nm, be 5.34seen nm,that 3.96the nm,average and 3.44sizes nm when times were min, 60 min, 80 min, and were 3.68 nm, 5.34 nm, 3.96 nm, and 3.44 nm when the maturing times were 40 min, 60 min, min, 100 min, respectively. Therefore, the average size of the CdS nanoparticles was at a maximum80when and 100 min, respectively. Therefore, the average size of the CdS nanoparticles was at a maximum the maturing time in the ultrasonic setup was 60 min. This was because the cavitation of the ultrasonic when the maturing time in setup wasi.e., 60 min. Thiscrystallization was because the of the treatment had two effects onthe theultrasonic CdS nanoparticles, induced andcavitation shearing. When ultrasonic treatment had two effects on the CdS nanoparticles, i.e., induced crystallization and shearing. the maturing time in the ultrasonic equipment was short (40 min), the growth of CdS grains was When the maturing timeof inthe the obtained ultrasonicCdS equipment was short min), thesmall. growth of CdS was incomplete, so the size nanoparticles was(40 relatively When thegrains maturing incomplete, so the size of the obtained CdS nanoparticles was relatively small. When the maturing time was 60 min, the CdS grains grew completely and the average particle size reached the maximum time was 60 min, the CdS grains completely andthe theCdS average particle size reached the due maximum value. If the maturing time wasgrew further prolonged, nanoparticles were broken to the value. If theshearing maturing timeofwas prolonged, theAs CdS nanoparticles wereparticle brokensize dueoftoCdS the continuous effect the further ultrasonic treatment. a result, the average continuous shearing effect of the ultrasonic treatment. As a result, the average particle size of CdS decreased. The studies of Dai et al. [17] showed the particle size of CdS nanoparticles with 60 mM 3decreased. The studies Dai etasal.a [17] showed the particle sizeItofwas CdSobserved nanoparticles with 60 mM mercaptopropionic acidof (MPA) function of incubation time. that the growth of 3-mercaptopropionic acid (MPA) as a function of incubation time. It was observed that the growth of the CdS particles rapidly increased at the initial stage and gradually slowed down, indicating the the CdS particles rapidly increased at the initial stage and gradually slowed down, indicating the MPA MPA inhibition on CdS particle growth. In this study, the size of CdS particles was related to the inhibition time on CdS growth.setup. In thisHowever, study, thethe sizeCdS of CdS particles was related to the maturing maturing in particle the ultrasonic nanoparticles had smaller sizes due to time in the ultrasonic setup. However, the CdS nanoparticles had smaller sizes due to shearing effect shearing effect of the ultrasonic when the maturing time was too long. The results were consistent of the ultrasonic when the maturing time was too long. The results were consistent with the PL spectra. with the PL spectra. In other words, the CNF/CdS nanocomposite showed only red when the In other words, the 60 CNF/CdS nanocomposite showed when the maturing time100 wasmin, 60 min. maturing time was min. If the maturing time was 40only min red or prolonged to 80 min and the If the maturing time was 40 min or prolonged to 80 min and 100 min, the size of the resultant CdS size of the resultant CdS nanoparticles decreased significantly, and the emission peak of the PL nanoparticles significantly, and the emission peakduring of the the PL preparation spectra showed anCNF/CdS obvious spectra showeddecreased an obvious blueshift phenomenon. Therefore, of the blueshift phenomenon. Therefore, during the preparation of the CNF/CdS nanocomposite material, nanocomposite material, the size of the CdS nanoparticles could be adjusted by controlling the the size oftime the CdS nanoparticles be adjusted by controlling the maturingcould time for ultrasonic maturing for the ultrasonic could maturing. Accordingly, the PL performance be the adjusted and maturing. Accordingly, the PL performance could be adjusted and its variation was very significant. its variation was very significant. Nanomaterials 2016, 6, 164

(1)

(2)

Figure TheThe size distributions of the of CdSthe nanoparticles with different and (2)time; the Figure 8.8.(1) (1) size distributions CdS nanoparticles with maturing different time; maturing TEM images of CNF/CdS nanocomposites with different maturing time; (a) 40 min; (b) 60 min; (c) 80 and (2) the TEM images of CNF/CdS nanocomposites with different maturing time; (a) 40 min; 2+ ratio 1.0, Na2S feeding min; and (d) 100 min (–COOH/Cd time 20 min). 2+ (b) 60 min; (c) 80 min; and (d) 100 min (–COOH/Cd ratio 1.0, Na S feeding time 20 min). 2

2.4. Effect of CdS Particle Size on CNF/CdS Nanocomposite PL Performance 2.4. Effect of CdS Particle Size on CNF/CdS Nanocomposite PL Performance The effect of CdS quantum dot size on the PL performance was analyzed according to the size The effect of CdS quantum dot size on the PL performance was analyzed according to the size of of the CdS particles (S) and the fluorescence emission wavelength (λ) of the CNF/CdS nanocomposite the CdS particles (S) and the fluorescence emission wavelength (λ) of the CNF/CdS nanocomposite material based on the above analysis data. The results are shown in Figure 9. When the size of the material based on the above analysis data. The results are shown in Figure 9. When the size of the CdS CdS particles increased from 3.44 nm to 5.34 nm, the fluorescence emission wavelength of the particles increased from 3.44 nm to 5.34 nm, the fluorescence emission wavelength of the CNF/CdS CNF/CdS nanocomposite gradually redshifted from 546 nm to 655 nm, and their fluorescence color changed from green, to yellow, to red. In addition, the larger the CdS size was, the more significant the redshift was. This was in agreement with the quantum confinement effect [22–24]. Based on the data in Figure 9, the size of the CdS particles (S) and the fluorescence emission wavelength (λ) of the

Nanomaterials 2016, 6, 164

9 of 12

nanocomposite gradually redshifted from 546 nm to 655 nm, and their fluorescence color changed from green, to yellow, to red. In addition, the larger the CdS size was, the more significant the redshift was. This was in2016, agreement with the quantum confinement effect [22–24]. Based on the data in Figure 9, Nanomaterials 6, 164 9 of 12 the size of the CdS particles (S) and the fluorescence emission wavelength (λ) of the composite material composite werebetween fitted. The relation between λ and Sas could be expressed follows (Equation were fitted.material The relation λ and S could be expressed follows (Equationas (2)). The correlation (2)). The correlation coefficient was 0.850.coefficient was 0.850. λ = 271 + 233 ln(S), (2) λ = 271 + 233 ln(S), (2) The nanocomposite was was decided The results results indicated indicated that that the the PL PL performance performance of of the theCNF/CdS CNF/CdS nanocomposite decided by by the size of CdS quantum dot. The size adjustable CdS quantum dot could be prepared by controlling the size of CdS quantum dot. The size adjustable CdS quantum dot could be prepared by controlling the –COOH/Cd2+2+ratio, ratio, the feeding time of Na maturing in ultrasonic. Therefore, 2 S and the –COOH/Cd the feeding time of Na 2S and the the maturing timetime in ultrasonic. Therefore, a PL aperformance PL performance tunable CNF/CdS nanocomposite could be obtained. tunable CNF/CdS nanocomposite could be obtained. (5.34, 655)

660

(4.86, 640)

640

(4.59,625)

Wavelength (nm)

620

(4.01, 613) (3.76, 605)

600

(3.96, 580)

580

(3.68, 560) 560

(3.44, 546) 540 520 3.0

3.5

4.0

4.5

5.0

5.5

Particle size (nm) (a)

(b) Figure 9. 9. (a) The effect of CdS quantum dot size on the PL PL performance; performance; and (b) the fluorescence fluorescence Figure CNF/CdS nanocomposites images of the CNF/CdS nanocomposites under under different different CdS quantum dot sizes.

3. Materials 3. Materialsand andMethods Methods 3.1. Materials The bleached Eucalyptus kraft pulp was donated by Aracruz Celulose (Espirito Santo, Brazil) with aa α-cellulose α-cellulosecontent contentofof86.8% 86.8%and andthe thedegree degreeofofpolymerization polymerization 1030. The other materials used with 1030. The other materials used in in this study their respective sources as follows: this study andand their respective sources areare as follows:     

CdCl2: Tianjin Fu Chen Chemical Reagents Factory, Tianjin, China, AR; Na2S: Guangzhou Chemical Reagent Factory, Guangzhou, China, AR; TEMPO: Alfa Aesar China, Shanghai, China; NaClO: 10%, Tianjin Fuyu Fine Chemical Co., Ltd., Tianjin, China; NaBr: Tianjin Kemiou Chemical Reagent Co., Ltd., Tianjin, China, AR.

Nanomaterials 2016, 6, 164

• • • • •

10 of 12

CdCl2 : Tianjin Fu Chen Chemical Reagents Factory, Tianjin, China, AR; Na2 S: Guangzhou Chemical Reagent Factory, Guangzhou, China, AR; TEMPO: Alfa Aesar China, Shanghai, China; NaClO: 10%, Tianjin Fuyu Fine Chemical Co., Ltd., Tianjin, China; NaBr: Tianjin Kemiou Chemical Reagent Co., Ltd., Tianjin, China, AR.

3.2. Methods 3.2.1. Preparation of CNFs The oxidized celluloses were prepared with bleached eucalyptus kraft pulp by oxidization in a TEMPO/NaBr/NaClO alkaline medium. The detailed preparation process is described in [25]. The oxidized celluloses with a carboxyl content of 1.63 mmol/g were dispersed in deionized water. The dispersion concentration was 0.3% (w/w). Continuous ultrasonic wave equipment (Guangzhou Newpower Ultrasonic Electronic Equipment Co. Ltd., Guangzhou, China) was used with a working period of 10 s and a pause period of 5 s. The CNF water suspension was obtained for future use. 3.2.2. Preparation of CNF/CdS Nanocomposite A CNF water suspension (100 g) (carboxyl content of 1.63 mmol/g, concentration of 0.3% (w/w)) was sampled. The amount of Cd2+ was calculated according to the –COOH/Cd2+ ratio. A CdCl2 solution (50 mL) of a certain concentration was added dropwise into the CNF water suspension under mechanical stirring in a water bath at 20 ◦ C. The adsorption time of the CdCl2 solution was controlled for 20 min. After adding the CdCl2 solution, the suspension was further stirred for 1 h. After that, the suspension was rapidly transferred into an ultrasonic instrument. Then, the suspension was diluted to 50 mL, followed by the addition of the same amount of Na2 S solution. The maturing time in the ultrasonic instrument was 40–100 min. The operation mode of continuous ultrasonic wave equipment was again a working time of 10 s and a pause of 5 s. The nanocomposite materials obtained under different reaction conditions were dialyzed for one week with deionized water and stored in the dark for future use. 3.2.3. TEM Analysis of CNF/CdS Nanocomposite The morphology of CNF/CdS nanocomposite material was analyzed by a TEM (H-7650 HITACHI Corporation, Tokyo, Japan). Between 80 and 100 CdS particles in the TEM image were selected for statistical analysis using the Nano Measurer software (Microsoft Corporation, Redmond, WA, USA) to obtain the nanoparticle size distribution. 3.2.4. PL Spectrum of CNF/CdS Nanocomposite The PL spectrum of the CNF/CdS nanocomposite was measured by a fluorescence spectrophotometer (F-7000 HITACHI High-Technologies Corporation, Tokyo, Japan). The excitation wavelength was 365 nm. 4. Conclusions The CNF/CdS nanocomposites with tunable photoluminescence were successfully prepared by an in situ uniformly compositing the quantized CdS nanoparticles onto CNFs. The CdS nanoparticles size and the emission wavelength of the CNF/CdS nanocomposite could be controlled by the following three ways. Firstly, by controlling the –COOH/Cd2+ ratio, the size of CdS nanoparticle could be changed from 4.01 nm to 5.34 nm, and the emission wavelength of the CNF/CdS nanocomposite varied from 613 nm to 655 nm. Secondly, by prolonging the feeding time of Na2 S from 10 min, 30 min to 40 min, the size of the CdS particles changed from 3.76 nm, to 4.86 nm, to 4.59 nm. The PL peak of the CNF/CdS nanocomposite showed a red shift, changing from 604 nm to 640 nm,

Nanomaterials 2016, 6, 164

11 of 12

then blueshifted to 625 nm. Thirdly, by changing the ultrasonic maturing times from 40 min to 100 min, the CdS nanoparticles size could be changed from 3.44 nm to 5.34 nm, and the fluorescence color of the CNF/CdS nanocomposite changed from green to red with the emission wavelength varying from 546 nm to 655 nm. Therefore, the size of the CdS nanoparticles could be controlled from 3.44 nm to 5.34 nm by adjusting the –COOH/Cd2+ ratio, the feeding time of Na2 S and the ultrasonic maturing time. As a result, the PL performance of the CNF/CdS nanocomposite could be adjusted gradually from 546 nm to 655 nm, realizing the PL tunable effect. The maturing time in ultrasonic conditions showed the most significant influence on the adjustment of PL performance. This CNF/CdS nanocomposite with a tunable PL wavelength has wide potential applications in biomarkers, anti-forgery ink, and photoelectric devices. Acknowledgments: This work was supported by State Key Laboratory of Pulp and Paper Engineering (2015C09, 201532), the Fundamental Research Funds for the Central Universities (2015ZM180), the financial support of the Guangdong-Hongkong joint innovation program (2014B050505019) and the National Natural Science Foundation of China (31570569). Author Contributions: Aimin Tang and Yuan Liu conceived and designed the experiments; Yuan Liu performed the experiments; Yuan Liu and Qinwen Wang analyzed the data; Zhiqiang Fang and Shiyu Fu contributed reagents and analysis tools; Qinwen Wang and Aimin Tang wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations The following abbreviations are used in this manuscript: CNFs TEM PL TEMPO MPA

Cellulose nanofibrils Transmission Electron Microscope Photoluminescence 2,2,6,6-tetramethylpiperidine-1-oxyradical 3-mercaptopropionic acid

References 1. 2. 3.

4.

5. 6. 7.

8.

9.

Li, X.; Chen, S.Y.; Hu, W.L.; Shi, S.K.; Shen, W.; Zhang, X.; Wang, H.P. In situ synthesis of CdS nanoparticles on bacterial cellulose nanofibers. Carbohyd. Polym. 2009, 76, 509–512. [CrossRef] Selvan, S.T.; Tan, T.T.Y.; Yi, D.K.; Jana, N.R. Functional and multifunctional nanoparticles for bioimaging and biosensing. Langmuir 2010, 26, 11631–11641. [CrossRef] [PubMed] Antony, J.V.; Kurian, P.; Vadakkedathu, N.P.N.; Kochimoolayil, G.E. In Situ Synthesis of CdS Quantum Dot–Partially Sulfonated Polystyrene Composite: Characterization and Optical Properties. Ind. Eng. Chem. Res. 2014, 53, 2261–2269. [CrossRef] Huang, D.W.; Niu, C.G.; Wang, X.Y.; Lv, X.X.; Zeng, G.M. “Turn-On” fluorescent sensor for Hg2+ based on single-stranded DNA functionalized Mn: CdS/ZnS quantum dots and gold nanoparticles by time-gated mode. Anal. Chem. 2013, 85, 1164–1170. [CrossRef] [PubMed] Chen, Y.F.; Rosenzweig, Z. Luminescent CdS quantum dots as selective ion probes. Anal. Chem. 2002, 74, 5132–5138. [CrossRef] [PubMed] Pattabi, M.; Amma, B.S. Synthesis and stability studies of thiophenol capped CdS nanoparticles. Sol. Energy Mat. Sol. Cells 2006, 90, 2377–2383. [CrossRef] Chen, D.A.; Zhao, F.; Qi, H.; Rutherford, M.; Peng, X.G. Bright and stable purple/blue emitting CdS/ZnS core/shell nanocrystals grown by thermal cycling using a single-source precursor. Chem. Mater. 2010, 22, 1437–1444. [CrossRef] Yang, J.Z.; Yu, J.W.; Fan, J.; Sun, D.P.; Tang, W.H.; Yang, X.J. Biotemplated preparation of CdS nanoparticles/bacterial cellulose hybrid nanofibers for photocatalysis application. J. Hazard. Mater. 2011, 189, 377–383. [CrossRef] [PubMed] Zeng, J.H.; Zhu, Y.; Liu, Y.F.; Yang, J.; Qian, Y.T.; Zheng, H.G. Morphology development of CdS/PVAc composite from spheres to rods. Mat. Sci. Eng. B 2002, 94, 131–135. [CrossRef]

Nanomaterials 2016, 6, 164

10. 11.

12. 13.

14. 15. 16.

17. 18. 19. 20. 21. 22.

23.

24.

25.

12 of 12

Qi, L.M.; Colfen, H.; Antonietti, M. Synthesis and characterization of CdS nanoparticles stabilized by double-hydrophilic block copolymers. Nano Lett. 2001, 1, 61–65. [CrossRef] Wang, S.M.; Liu, P.; Wang, X.X.; Fu, X.Z. Homogeneously distributed CdS nanoparticles in nafion membranes: Preparation, characterization, and photocatalytic properties. Langmuir 2005, 21, 11969–11973. [CrossRef] [PubMed] Han, J.; Su, H.L.; Dong, Q.; Zhang, D.; Ma, X.X.; Zhang, C.F. Patterning and photoluminescence of CdS nanocrystallites on silk fibroin fiber. J. Nanopart. Res. 2010, 12, 347–356. [CrossRef] Nguyen, T.D.; Hamad, W.Y.; MacLachlan, M.J. CdS Quantum Dots Encapsulated in Chiral Nematic Mesoporous Silica: New Iridescent and Luminescent Materials. Adv. Funct. Mater. 2014, 24, 777–783. [CrossRef] Pinto, R.J.B.; Carlos, L.D.; Marques, P.A.A.P.; Silvestre, A.J.D.; Freire, C.S.R. An overview of luminescent bio-based composites. J. Appl. Polym. Sci. 2014, 131. [CrossRef] Hu, W.L.; Chen, S.Y.; Yang, J.X.; Li, Z.; Wang, H.P. Functionalized bacterial cellulose derivatives and nanocomposites. Carbohyd. Polym. 2014, 101, 1043–1060. [CrossRef] [PubMed] Lokanathan, A.R.; Uddin, K.M.A.; Rojas, O.J.; Laine, J. Cellulose nanocrystal-mediated synthesis of silver nanoparticles: Role of sulfate groups in nucleation phenomena. Biomacromolecules 2013, 15, 373–379. [CrossRef] [PubMed] Dai, J.; Yang, X.Y.; Hamon, M.; Kong, L.Z. Particle size controlled synthesis of CdS nanoparticles on a microfluidic chip. Chem. Eng. J. 2015, 280, 385–390. [CrossRef] Ruan, D.; Huang, Q.L.; Zhang, L.N. Structure and properties of CdS/regenerated cellulose nanocomposites. Macromol. Mater. Eng. 2005, 290, 1017–1024. [CrossRef] Martinez-Castanon, G.; Loyola-Rodriguez, J.; Reyes-Macias, J.; Nino-Martinez, N.; Ruiz, F. Synthesis and optical properties of functionalized CdS nanoparticles with different sizes. Superf. Vacio 2010, 23, 1–4. Pattabi, M.; Amma, B.S.; Manzoor, K. Photoluminescence study of PVP capped CdS nanoparticles embedded in PVA matrix. Mater. Res. Bull. 2007, 42, 828–835. [CrossRef] Brus, L. Electronic wave functions in semiconductor clusters: Experiment and theory. J. Phy. Chem. 1986, 90, 2555–2560. [CrossRef] Shombe, G.B.; Mubofu, E.B.; Mlowe, S.; Revaprasadu, N. Synthesis and characterization of castor oil and ricinoleic acid capped CdS nanoparticles using single source precursors. Mat. Sci. Semicond. Proc. 2016, 43, 230–237. [CrossRef] Osuntokun, J.; Ajibade, P.A. Synthesis and structural characterization of CdS nanoparticles using nitrogen adducts of mixed diisopropylthiourea and dithiolate derivatives of Cd(II) complexes. Superlattices Microstruct. 2015, 83, 89–100. [CrossRef] Rodriguez-Castaneda, C.A.; Moreno-Romero, P.M.; Martinez-Alonso, C.; Hu, H.L. Microwave Synthesized Monodisperse CdS Spheres of Different Size and Color for Solar Cell Applications. J. Nanomater. 2015, 2015. [CrossRef] Isogai, A.; Kato, Y. Preparation of polyglucuronic acid from cellulose by TEMPO-mediated oxidation. Cellulose 1998, 5, 153–164. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).