Analytical Science and Technology (분석과학)

The Korean Society of Analytical Sciences (한국분석과학회)
권호 표지
DOI QR코드

DOI QR Code

[Retraction]Characterization of carbon black nanoparticles using asymmetrical flow field-flow fractionation (AsFlFFF)

  • Received : 2018.11.30
  • Accepted : 2019.05.07
  • Published : 2019.06.25
Download PDF
⟨ Previous Next ⟩

Abstract

High viscosity carbon black dispersions are used in various industrial fields such as color cosmetics, rubber, tire, plastic and color filter ink. However, carbon black particles are unstable to heat due to inherent characteristics, and it is very difficult to keep the quality of the product constant due to agglomeration of particles. In general, particle size analysis is performed by dynamic light scattering (DLS) during the dispersion process in order to select the optimum dispersant in the carbon black dispersion process. However, the existing low viscosity analysis provides reproducible particle distribution analysis results, but it is difficult to select the optimum dispersant because it is difficult to analyze the reproducible particle distribution at high viscosity. In this study, dynamic light scattering (DLS) and asymmetrical flow field-flow fractionation (AsFlFFF) analysis methods were compared for reproducible particle size analysis of high viscosity carbon black. First, the stability of carbon black dispersion was investigated by particle size analysis by DLS and AsFlFFF according to milling time, and the validity of analytical method for the selection of the optimum dispersant useful for carbon black dispersion was confirmed. The correlation between color and particle size of particles in high viscosity carbon black dispersion was investigated by using colorimeter. The particle size distribution from AsFlFFF was consistent with the colorimetric results. As a result, the correlation between AsFlFFF and colorimetric results confirmed the possibility of a strong analytical method for determining the appropriate dispersant and milling time in high viscosity carbon black dispersions. In addition, for nanoparticles with relatively broad particle size distributions such as carbon black, AsFlFFF has been found to provide a more accurate particle size distribution than DLS. This is because AsFlFFF, unlike DLS, can analyze each fraction by separating particles by size.

Keywords

Introduction

Carbon black is a color pigment with high blackness, and because of its excellent colorability, electrical conductivity, weather resistance, and chemical resistance, it has broad range of applications, including as a reinforcing agent and filler for plastics andelastomers1 and a shading agent for a black matrix.2-4However, carbon black has limitations such as a very small primary particle diameter, high porosity, alarge specific surface area, and a strong tendency t oform agglomerates owing to strong affinity for eachother, which results in its poor dispersion stability. Moreover, as carbon black is hydrophobic, it haspoor wettability for water, making it very difficult todisperse high concentrations in aqueous systems.5-8

An effective method for dispersing pigment particlesis to use the electrostatic repulsion of electrically charged particles dispersed in a solution.4,9,10Alternatively, an adsorbed surface layer can providesteric hindrance11-14 to maintain a certain spacing between the pigment particles in order to prevent them from agglomerating.15,16 Fig. 1 shows the three basic stages of pigment dispersion.

Stage 1 (Wetting): This process involves removingair and moisture present on the surface of the pigmentand replacing them with a solution. Here, the solution penetrates into the space between agglomerates.

Stage 2 (Grinding): In this stage, the pigmentagglomerates are ground to appropriate size. The particles become smaller in size as the pigmentagglomerates are ground by mechanical force (impactand shear force). As the particles become smaller, the surface area of the pigment increases to createinstability.

Stage 3 (Stabilization): In this stage, precipitation and re-agglomeration of dispersed particles areprevented. The force that prevents agglomerationoriginates either from electrostatic repulsion betweencharged particles or the steric hindrance effect of apolymer adsorption layer.17,18

Fig. 2 shows a simple depiction of three components in carbon black dispersion. For stable dispersion, it isimportant to achieve a balance between the three components of dispersion: pigment, solvent, and dispers ant. Among these three components, the acid-base affinity between the carbon black particles and the dispersant is especially important.19,20

Generally, dispersants are divided into low molecular weight (LMW) and high molecular weight (HMW) types. LMW dispersants typically lack effectivenesson organic pigments consisting of carbon, hydrogen, oxygen, and nitrogen atoms. Accordingly, HMWdispers ants are generally used and they usually havean adsorption group with a chain that can be adsorbed on the surface of an organic pigment particle.10

In the present study, mechanical disruption21 was used to analyze the size of particles in a carbon black dispersion according to the type of dispersant and dispersion time. Moreover, styrene maleic acid (SMA) dispersant, a HMW dispersant with excellent thermal stability, was used to prepare the carbon black dispersion. The thermal stability at high temperatures was assessed to verify the stability of the prepared black carbon dispersion. In addition, the study investigated the feasibility of an analytical methodusing scanning electron microscopy (SEM), dynamiclight scattering (DLS), and asymmetrical flow field-flow fractionation (AsFlFFF)22 for particle size analysis to determine the appropriate dispersant and dispersion time for carbon black particles.

 

BGHHBN_2019_v32n3_77_f0001.png 이미지

Fig. 1. Three stages of pigment dispersion.

 

BGHHBN_2019_v32n3_77_f0002.png 이미지

Fig. 2. Three-component system of carbon black dispersion.

 

2. Theory of AsFlFFF

Field-flow fractionation is a system that separates materials based on the differences in the diffusion coefficient of materials passing through a hollowchannel. An AsFlFFF channel separates particles by releasing the fluid into space created by placing aspacer in between two plates. Because of the structural characteristics of the channel, a parabolic flow is formed, where the flow rate in both sides of the plates is slow owing to the surface tension formedinside the channel with the flow rate increasing toward the center of the plate. Consequently, elutionoccurs in the order of smaller particles to larger particles over time, whereby separation by particlesize also occurs. By altering the rate of cross-flowthat acts perpendicularly to the channel over time, the sample retention time could be adjusted to increasethe separation capability (field-programming).23

One of the key features of AsFlFFF (AF4) is that the hydrodynamic diameter (dH) can be calculated directly by measuring the sample retention time (tr) (refer to Eq. (1)).24-26

\(d_{H}=\frac{2 k T V^{0}}{\pi \eta w^{2} V_{c} t^{0}} t_{r}\)       (1)

In Eq. (1), k is the Boltzmann constant; T is the absolute temperature (K); V0 is the void volume of the channel; η is the viscosity of the carrier liquid; w is the channel thickness; Vc is the cross-flow rate; and t0 is the time required to pass through the channelvolume (void time). In Eq. (1), all the variables, except tr, are constants under the given experimental conditions, and the tr of the sample can be measured by AF4 to determine the particle size and sizedistribution of the sample directly.19,27,28

 

3. Experimental

 

3.1. Materials

Carbon black (S.A. 75 m2/g, Bulk density 80-120 g/L, Alfa Aesar, USA) was used to prepare the carbon black dispersion. For the synthesis of the dispers ant, deionized water (Milli Q PLUS), SMA-1000 (styrenemaleic anhydride 1000), SMA-2000 (styrene maleicanhydride 2000), and SMA-3000 (styrene maleicanhydride 3000, SARTOMER Co. Ltd, Pennsylvania, USA) were used, along with ammonia water (25~28%, DUKSAN, KOREA) for acidification. Moreover, FL-70 (Fisher Chemical, New Jersey, USA) and sodium azide (NaN3, Sigma-Aldrich, St. Louis, USA) were used to prepare the AF4 carrier liquid.

 

3.2. Analytical instruments

A digital overhead stirrer (HT-50DX, DAIHAN-brand, Korea) was used to prepare the carbon black dispersion, and a heating mantle (DH.WHM 12214, DAIHAN-brand, Korea) was used to apply heat during stirring. The physical properties and stability of the prepared carbon black solution were assessed using a pH and conductivity meter (OHAUS Corp, OHAUS STARTER 3100C, USA) and a viscosity meter (Brookfield, LVDVE 230, USA). To measurethe color of carbon black, a colorimeter (CR-400 Chroma meter, Konica Minolta, USA) was used.

Moreover, a transmission electron microscope (TEM; JEM-F200, JEOL Ltd., Japan) and field-emissionscanning electron microscope (FE-SEM; JEOL-7800F, JEOL Ltd., Japan) were used for the identification of the morphology and size of the carbon black particles, whereas dynamic light scattering (DLS; ELSZ-2000, Otsuka, Japan), AF4 short channel (Wyatt Tech, Europe GmbH, Dernbach, Germany), cellulose membrane with cut-off MW of 10 kDa (Millipore, Bedford, USA), and a Mylar space with the thickness of 250 µm were used for the measurement of particle size and size distribution. The carrier liquid used in AF4 analysis was an aqueous solution containing 0.1% FL-70 and 0.02% NaN3. The flow of the carrierliquid was generated using a high-performance liquid chromatography (HPLC) pump (P-6000, FUTECSCo., Ltd, Korea), whereas an Optiflow 1000 Liquid Flow meter (Agilent Technologies, Palo Alto, CA, USA) was used to measure the flow rate. A UV detector (Spectra SERIES UV 150, THERMO SEPARATION PRODUCTS, USA) was used to detect the sampleseluted after being separated by particle size from AF4. The analysis was performed with a channel flow of 0.8 mL/min and cross-flow of 0.3 mL/min. The samples were injected using a syringe pump(Legato 110, KD Scientific Inc., Mendon, USA), with 50 µL being injected at a rate of 0.2 mL/min. To evaluate the reproducibility of the analysis results, all the measurements were repeated thrice.

 

3.3. Synthesis of dispersant

For the dispersion of carbon black particles, three types of dispersants were synthesized.29 First, afteradding 42 g of SMA powder to a three-neck flask, 42 g of DI water was added and mixed by stirring. Here, a heating mantle was used to raise the temperature of the solution up to 80 °C. Once the temperature of the solution reached 80 °C, 24 g of ammonia water was added and mixed by stirring for 2 h to synthesize the dispersant. As there were three different types of SMA powder, three different types of dispersant wereprepared. Fig. 3 shows the structure of the synthesized SMA 1000, SMA 2000, and SMA 3000.

 

BGHHBN_2019_v32n3_77_f0003.png 이미지

Fig. 3. Structure of synthesized SMA dispersant. (a) SMA1000, (b) SMA 2000, (c) SMA 3000.

 

3.4. Preparation of high-viscosity carbon black dispersion

For the preparation of carbon black dispersion, abasket mill (HSX0502251, Hyosung, Korea) was used. Accordingly, 75 g of SMA dispersant and 420g of distilled water were added to 5 g of carbon black. The size of the beads used was 0.8 mm and atotal of 350 g of beads were used to performmilling at 1,500 rpm.

 

4. Results and Discussion

 

4.1. Storage stability analysis of carbon black dispersion

High-viscosity carbon black dispersion requires anaccurate size analysis method for the optimization of dispersion time and dispersant. DLS, which iscurrently used, shows reproducibility for the particledistribution of a low-viscosity dispersion, but it has the disadvantage of not being able to producereproducible results with a high-viscosity dispersionowing to the decline in particle mobility. Owing tosuch a reason, high-viscosity carbon black dispersionshave limited industrial application. In the present study, different types of dispersants and milling times were used to assess storage stability over 11 days at 60 °C. Figs. 4, 5, and 6 show the changes in pH, viscosity, conductivity, and particle size over time in the carbon black dispersion prepared using SMA1000, SMA 2000, and SMA 3000 dispers ants, respectively. In all the cases, the milling time was 1-3 h. It was confirmed that, after 3 h, the particles became bigger owing to overdispersion. For the confirmation of stability, measurements were obtained over 11 days at 60 °C. It was also confirmed that most cases showed slight changes after 11 days. Theresults shown in Figs. 4, 5, and 6 are summarized in Tables 1, 2, and 3, respectively. All the subsequentexperiments showed the results from three repeated measurements.

 

BGHHBN_2019_v32n3_77_f0004.png 이미지

Fig. 4. The pH change of carbon black dispersion prepared by using SMA 1000 according to milling time (1~3 hr) at 60 o C, 11 days (a), viscosity (b), conductivity (c), particle size (d).

 

BGHHBN_2019_v32n3_77_f0010.png 이미지

Fig. 5. The pH change of carbon black dispersion prepared by using SMA 2000 according to milling time (1~3 hr) at 60 o C, 11days (a), viscosity (b), conductivity (c), particle size (d).

 

BGHHBN_2019_v32n3_77_f0006.png 이미지

Fig. 6. The pH change of carbon black dispersion prepared by using SMA 3000 according to milling time (1~3 hr) at 60 oC, 11 days (a), viscosity (b), conductivity (c), particle size (d).

 

Table 1. The storage stability of carbon black dispersion prepared by using SMA 1000 according to milling time (1~3 hr) at 60 o C, 11 days

BGHHBN_2019_v32n3_77_t0002.png 이미지

 

Table 2. The storage stability of carbon black dispersion prepared by using SMA 2000 according to milling time (1~3 hr) at 60 o C, 11 days

BGHHBN_2019_v32n3_77_t0003.png 이미지

 

Table 3. The storage stability of carbon black dispersion prepared by using SMA 3000 according to milling time (1~3 hr) at 60 o C, 11 days

BGHHBN_2019_v32n3_77_t0004.png 이미지

Figs. 4, 5, and 6 and Tables 1, 2, and 3 showed that the pH and viscosity remained constant without significant change. However, the carbon black dispersion prepared using SMA 1000 dispersant showed the largest decrease in conductivity i.e., 77.4 % on day 11, as compared with the baseline. This decrease was larger than that observed when other dispers ants were used. It is suspected that the surface chargevalues of the particles gradually decreased owing to the cohesive force between the particles. Moreover, the 2 h dispersion data of SMA 2000 and SMA 3000 remained consistent over 11 days, which indicated that dispersion for 2 h had the highest stability.

 

4.2. Morphology of carbon black particles

SEM images were examined to check the morphology of the prepared carbon black particles.

An examination of the SEM images showed that the primary particle diameter of carbon black was within the range of 30~40 nm. Moreover, the particles were present as irregularly shaped agglomerates, but not spherical ones. These results demonstrated that the dispersion process stably maintained the appearance of the carbon black particles withoutany deformation.

 

BGHHBN_2019_v32n3_77_f0007.png 이미지

Fig. 7. SEM image of carbon black particles according to dispersant and milling time (X 100,000). (a) SMA 1000, milling 1 hr, (b) SMA 1000, milling 2 hr, (c) SMA 1000, milling 3 hr, (d) SMA 2000, milling 1 hr, (e) SMA 2000, milling 2 hr, (f) SMA 2000, milling 3 hr, (g) SMA 3000, milling 1 hr, (h) SMA 3000, milling 2 hr, (i) SMA 3000, milling 3 hr.

 

4.3. Particle size analysis of carbon black dispersion

Particle size analysis of the prepared carbon black dispersion was performed using DLS and AF4. The sample analyzed was carbon black dispersion stored for 11 days at 60 °C. Fig. 8 and Table 4 show the results from DLS.

 

Table 4. DLS particle size analysis of carbon black suspensions

table 4..PNG 이미지

As shown in Fig. 8, the DLS particle size analysis results showed the progression of agglomerationowing to overdispersion after 3 h of milling, regardless of the type of dispersant. Moreover, based on the appearance of two peaks after 1 h of milling, anundispersed state was suspected. Accordingly, it was determined that 2 h of milling was the optimal dispersion time. Considering the a forementioned conductivity results and DLS results, a stable dispersion of carbon black particles may be difficult to achieve with the SMA 1000 dispersant. It is suspected that this is because the SMA 1000 dispersant is desorbed from the surface of carbon black because of the heatowing to the lack of styrene groups within the SMA1000 dispersant covalently bonded to carbon black. However, the DLS results showed that, as the range of error was too large, it was still difficult to determinethe optimal dispersant and dispersion time.

 

BGHHBN_2019_v32n3_77_f0008.png 이미지

Fig. 8. DLS particle size analysis of carbon black suspension according to milling time Milling. SMA 1000 (a), SMA 2000 (b), SMA 3000 (c).

Fig. 9 and Table 5 show the AF4 results for the same sample.

Unlike the DLS results, the AF4 measurementresults showed a tendency of gradual decrease in the particle size according to an increase in milling time when the SMA 1000 and SMA 2000 dispersants were used, whereas the particle size tended to increaseaccording to the milling time when the SMA 3000 dispers ant was used. As with the DLS results, thisindicated that, when SMA 3000 was used, 3 h of dispersion resulted in a gradual increase in the particle size owing to overdispersion. As shown in Table 5, the overall reproducibility of the AF4 particlesize analysis appeared more consistent than that of DLS. Especially, when SMA 3000 is compared, AF4 showed a relative standard deviation (SD) of 34.8, whereas DLS showed a relative SD of 175, which was approximately 5 times higher.

 

BGHHBN_2019_v32n3_77_f0009.png 이미지

Fig. 9. AF4 fractograms of carbon black suspension according to milling time. SMA 1000 (a), SMA 2000 (b), SMA 3000 (c). The AF4 channel flow rate was 0.3 mL/min and the cross flow rates was 0.3 mL/min, and the carrier liquid was water containing 0.1% FL-70 and 0.02% NaN

 

Table 5. AF4 particle size analysis of carbon black suspensions

Table 5.PNG 이미지

 

4.4. Colorimetric analysis of carbon black dispersion

A colorimeter was used for the additional analysis of carbon black dispersion. A colorimeter providesthe numerical values for brightness (L*), redness (a*), and yellowness (b*). Colorimetric measurements were performed using the stable dispersants SMA2000 and SMA 3000. Tables 6 and 7 show the colorimetric measurement results of the carbon black dispersion obtained using the SMA 2000 and 3000 dispers ants for the same samples as mentioned above (samples store for 11 days at 60 °C).

 

Table 6. Colorimetric results of carbon black suspensions prepared using SMA 2000

Table 6.PNG 이미지

 

Table 7. Colorimetric results of carbon black suspensions prepared using SMA 3000Table 7.PNG 이미지

Colorimetric measurement results showed that, when the SMA 2000 dispersant was used, the lowestL* value was observed at the milling time of 3 h, whereas when the SMA 3000 dispersant was used, the lowest L* value was observed at the milling time of 1 h. The L* value represents brightness and luminosity, where a lower L* value is closer to black, indicating the smallest particle size. The results in Tables 6 and 7 showed similarity with the AF4 results shown in Table 5 with respect to the correlations of particle size.

Based on these results, it is believed that additionaloptimization studies are required to use the SMA1000 dispersant for the dispersion of carbon black. AF4 and colorimetric analyses results confirmed that the preparation of stable carbon black dispersions would be possible with a milling time of 3 h when the SMA 2000 dispersant is used and 1 h when the SMA 3000 dispersant is used.

 

5. Conclusions

The present study analyzed the characteristics of carbon black particles. Carbon black tends to agglomerate easily due to low thermal stability. In particular, the agglomeration of carbon black used as a pigment can cause color changes and product defects. Therefore, the selection of a stable dispers antand optimal dispersion time is important. In the present study, appropriate types of dispersant and milling times for the preparation of stable carbon black dispersions were identified through AF4 and colorimetric analyses.

In the particle size analysis of high-viscosity carbon black dispersions, conventional DLS results showed relatively low reproducibility, whereas AF4 results showed relatively high reproducibility. Moreover, the correlations of AF4 and colorimetric analyses results for the type of dispersant and milling timeverified these as powerful analytical methods fordetermining the appropriate dispersant and milling time for high-viscosity carbon black dispersion.

 

Acknowledgements

This work was supported by 2019 Hannam University Research Fund and Korea Environment Industry & Technology Institute (KEITI) through Technology Development Program for Environmental Industry Advancement, funded by Korea Ministry of Environment (MOE) (RE201805141).

References

  1. B. B. Boonstra, and A. I. Medalia, Rubber Chemistry and Technology, 36(1), 115-142 (1963). https://doi.org/10.5254/1.3539530
  2. A. I. Medalia, Rubber Chemistry and Technology, 59(3), 432-454 (1986). https://doi.org/10.5254/1.3538209
  3. E. Auer, A. Freund, J. Pietsch, and T. Tacke, Appl. Catal. A Gen., 173(2), 259-271 (1998). https://doi.org/10.1016/S0926-860X(98)00184-7
  4. Y. Lin, T. W. Smith, and P. Alexandridis, Langmuir, 18(16), 6147-6158 (2002). https://doi.org/10.1021/la011671t
  5. K. Loganathan, D. Bose, and D. Weinkauf, Int. J. Hydrogen Energy, 39(28), 15766-15771 (2014). https://doi.org/10.1016/j.ijhydene.2014.07.125
  6. S. M. Iveson, J. D. Litster, K. Hapgood, and B. J. Ennis, Powder Technol., 117(1-2), 3-39 (2001). https://doi.org/10.1016/S0032-5910(01)00313-8
  7. F. Tiarks, K. Landfester, and M. Antonietti, Macromol. Chem. Phys., 202(1), 51-60 (2001). https://doi.org/10.1002/1521-3935(20010101)202:1<51::AID-MACP51>3.0.CO;2-J
  8. S. Lee, C. H. Eum, and W. J. Kim, Bull. Korean Chem. Soc., 60(4), 286-291 (2016). https://doi.org/10.5012/jkcs.2016.60.4.286
  9. P. A. Hartley and G. D. Parfitt, Langmuir, 1(6), 651-657 (1985). https://doi.org/10.1021/la00066a003
  10. M. Hermansson, Colloids Surf. B Biointerfaces, 14(1-4), 105-119 (1999). https://doi.org/10.1016/S0927-7765(99)00029-6
  11. R. J. Pugh, T. Matsunaga, and F. M. Fowkes, Colloids Surf., 7(3), 183-207 (1983). https://doi.org/10.1016/0166-6622(83)80046-8
  12. J. A. Belmont, J. E. Johnson, and C. E. Adams, US Patent number 5,571,311 (1996).
  13. C. E. Adams and J. A. Belmont, US patent number 5,895,522(1999).
  14. R. S. Whitehouse, US Patent number 5,872,177(1999).
  15. C. F. Lee, C. C. Yang, L. Y. Wang, and W. Y. Chiu, Polym., 46(15), 5514-5523 (2005). https://doi.org/10.1016/j.polymer.2005.04.053
  16. E. M. Dannenberg, Rubber Chemistry and Technology, 25(4), 843-857 (1952). https://doi.org/10.5254/1.3543445
  17. A. A. Tracton, CRC Press, 1st Edition, 528 (2006).
  18. T. Oyanagi and K. Nakano, US Patent number 7,763,108 (2010).
  19. W. Kim, J. Bae, C. H. Eum, J. Jung, and S. Lee, Microchem. J., 142, 167-174 (2018). https://doi.org/10.1016/j.microc.2018.06.035
  20. J. Bae, J. Jung, S. Lee, and W. Kim, J. Korean Oil. Chemists' Soc, 34(2), 357-366 (2017).
  21. T. Shibata, US Patent number 6,374,244 (2002).
  22. J. Bae, W. Kim, K. Rah, E. C. Jung, and S. Lee, Microchem. J., 104, 44-48 (2012). https://doi.org/10.1016/j.microc.2012.04.007
  23. H. Dou, E. C. Jung, and S. Lee, J. Chromatogr. A, 1393, 115-121 (2015). https://doi.org/10.1016/j.chroma.2015.03.025
  24. B. Wittgren, K. G. Wahlund, H. Derand, and B. Wesslen, Macromolecules, 29(1), 268-276 (1996). https://doi.org/10.1021/ma950837s
  25. J. C. Giddings, J. Sep. Sci., 1(1), 123-125 (1966).
  26. K. G. Wahlund, and J. C. Giddings, Anal. Chem., 59(9), 1332-1339 (1987). https://doi.org/10.1021/ac00136a016
  27. A. Litzen, and K. G. Wahlund, Anal. Chem., 63(10), 1001-1007 (1991). https://doi.org/10.1021/ac00010a013
  28. A. Litzen, Anal. Chem., 65(4), 461-470 (1993). https://doi.org/10.1021/ac00052a025
  29. D. Braun, R. Sauerwein, and G. P. Hellmann, Macromol. Symp., 163(1), 59-66 (2001). https://doi.org/10.1002/1521-3900(200101)163:1<59::AID-MASY59>3.0.CO;2-3