Carbon Nanotube and Graphene Quantification Kyle Doudrick SRC TeleSeminar July 11 th 213 1
Current Methods Available Thermogravimetric analysis Only accounts for weight change Radiolabeling Must be pre-labled Very sensitive Fluorescence Only valid for semiconducting SWCNTs Electrophoresis Sensitive Anything dark in color may interfere Not all CNTs will stay at gel interface Metal analysis (ICP) Only valid for CNTs containing metals Sensitive 2
Method Development Doudrick, K., P. Herckes, P. Westerhoff. Detection of carbon nanotubes in environmental matrices using programmed thermal analysis. Environ. Sci. Technol., 46(22), 12246 12253, 212. 3
Instrument used to developed analytical method Sunset Laboratory (Forest Grove, OR) Thermal Optical Transmittance Traditionally used to measure soot in air using NIOSH standards (Method 54B) and also organic (OC) and elemental carbon (EC) in air pollution studies 4
Sample Filters Sample Holder 5
Thermogram and carbon definitions Sucrose 6
CNT Characterization CNT ID CNT Type State Purity a Metal Content b Outer Diameter (nm) Inner Diameter (nm) Length (µm) MW-O MWCNT Raw >95% <6% 2-3 5-1 1-3 MW-P MWCNT Purified >98% <2% 2-3 5-1 1-3 MW-F MWCNT Functionalized >99.9% <.1% 2-3 5-1 1-3 MW-15 MWCNT Raw >95% <5% 7-15 3-6.5-2 MW-2 MWCNT Raw >95% <5% 1-2 5-1.5-2 MW-3 MWCNT Raw >95% <5% 1-3 5-1.5-5 MW-1 MWCNT Raw >95% <5% 6-1 5-1.5-5 MW-OH MWCNT Functionalized >95% <1.5% 8-15 3-5 1-5 MW-COOH MWCNT Functionalized >95% <1.5% 8-15 3-5 1-5 MW-15G c MWCNT Annealed >97% <1% 7-15 3-6.5-2 MW-Mitsui MWCNT Raw >98% <1% 2-7 NA NA MW-Arc MWCNT d Raw <5% % 5-1 e NA NA SW SWCNT Raw <5% <1% 1.1 NA.5-1 SW-65 SWCNT Purified <75% <1%.8 NA.45-2 a CNT content reported by manufacturer. MW-P and MW-F calculated assuming no amorphous carbon remaining. b Metal content reported by manufacturer except for MW-F and MW-P determined using energy dispersive X-ray spectroscopy and MW-15G using thermogravimetric analysis. c MW-15 annealed at ~2 C in UHP He. d Synthesized using arc method; all others are CVD. e Obtained from TEM images; all others reported by manufacturer. 7
CNT Thermograms Temperature ( C) 1 9 8 7 6 5 4 3 2 1 Temperature MW-O MW-P MW-F MW-15 MW-2 MW-3 MW-1 MW-OH MW-COOH MW-15G MW-Mitsui MW-Arc SW 14 12 1 8 6 4 2 FID Signal (for CNTs) 15 125 145 165 185 25 225 245 265 285 Time (seconds) 8
Method Development Inert Conditions Using NIOSH 54B method, some CNTs desorb/oxidize at higher temperatures These are classified as weak, while all that withstand higher temps are strong Percent CNT Mass Remaining 1% 8% 6% 4% 2% % MW-Arc MW-15 MW-F MW-O SW-65 MW-P Temperature 2 4 6 8 1 Time (seconds) 9 8 7 6 5 4 3 2 1 Temperature ( C) Oxidizing Conditions Some CNTs oxidize very early on at low temps These are classified as weak, while all that withstand higher temps are strong Percent CNT Mass Remaining 1% 8% 6% 4% 2% % SW-65 SW MW-F MW-O MW-15 MW-15G MW-Mitsui MW-Arc 5 1 15 2 Time (seconds) 1 9 8 7 6 5 4 3 2 1 9 Temperature ( C)
Method Development Inert Conditions Examined different maximum temperature conditions for a representative weak CNT 675 C was the max temp where no CNT loss occurred 12% Percent CNT Mass Remaining 1% 8% 6% 4% 2% % MW-O-675 MW-O-7 MW-O-75 MW-O-87 2 4 6 Time (seconds) 1
Can we measure CNTs Directly (w/o extraction)? Percent Mass Remaining 12% 1% 8% 6% 4% 2% Milk Human serum Urban air Sediment MW-F MW-15 MW-Arc Temperature 9 8 7 6 5 4 3 2 1 Temperature ( C) % 1 2 3 4 Time (seconds) At ~95% weak CNT mass remaining, there is still 7% urban air, 6% sediment, 5% serum, and 3% milk. Sediments are the most challenging with <1% interference with even the strongest CNT 11
Matrix Interference NEC: Non-CNT Elemental Carbon (i.e., soot, PEC) Simple: NEC < 1% Complex: NEC 1% Percent NEC in Sample 5% 4% 3% 2% 1% Simple Complex % 12
Using Raman to Determine CNT Thermal Classification 95 85 D-band G-band Intesnity 75 MW-P MW-Arc 65 55 45 35 5 75 112515175222525 Raman Shift (cm-1) Intensity 2 16 12 8 SW MW-P 4 1 15 2 25 3 35 4 45 5 Raman Shift (cm-1) 13
CNT Thermal Classification 2. I D /I G Ratio 1.6 1.2.8.4 Weak R 2 =.96 I G Strong I D. 6 7 8 9 1 Oxidation Temperature at 5% Mass Loss ( C) 14
Extraction 15
Extraction is key CNT Recovery Eight different reagents (acids, alkalis, enzymes) 6 C for 24 hrs with mixing Centrifugal separation with water washing in between Quantified using PTA 1% CNT Percent Recovery 8% 6% 4% 2% S-CNT W-CNT % 16
How to Compare Reagents and Analyze Damage to CNTs? Thermograms are overly complicated! FID Signal 1 8 6 4 2 3 4 5 Untreated Water Solvable Hydroxide Nitric Sulfuric Hydrochloric Hydrofluoric 1 Peroxide 3 6 Proteinase K 2 2 F-CNT 1 1 Temperature 2 4 6 8 Time (s) 9 8 7 6 5 4 Temperature ( C) 17
Thermogram Analysis Weak CNTs Fraction of Total Peak Area.5.4.3.2.1 4 35 3 25 2 Time to 5% Mass Loss (s) 5 C 55 C 6 C 65 C 7 C 75 C Time to 5% Solvable, HCl, HF, and pro K showed no change (<5%) HNO 3 had a 5-1% change Water, NaOH, H 2 O 2, and H 2 SO 4 had a 1-2% change Functionalized CNTs had a 4% decrease 18
Thermogram Analysis Strong CNTs Fraction of Total Peak Area 1..8.6.4.2. 75 7 65 6 Time to 5% Mass Loss (s) 7 C 75 C 8 C 91 C Time to 5% Solvable, HF, HNO 3, H 2 O 2, NaOH, and H 2 SO 4, and pro K showed no change (<5%) Water and HCl had a 5-1% change So HNO 3, HCl, and HF should be okay to use if separation method can be improved 19
Separation Filtration was only optimal for CNTs that were aggregated Functionalized or fully dispersed CNTs passed partly through the filter Filtration does not allow for washing of sample to remove interferences 2
Reagent selection Instrument Damage Possibly residual acids cause corrosion Additional washing steps could be incorporated into method, but this may reduce recovery 21
Application Cyanobacteria I D /I G =.26 ±.8 Cyanobacteria was a complex matrix and pretreatment was necessary Solvable or HNO 3 was adequate to dissolve CB Raman revealed the CNT to correctly to a strong CNT CNT Concentration, µg CNT/g CB (CNT mass, µg) Recovery 1 (.51) 16 ± 29% 54 (2.7) 99 ± 1.9% 22 (11) 96 ± 3.% Temperature (C) 1 8 6 4 2 Inert Oxidizing CNTs 2 4 6 8 1 8 6 4 2 FID Signal Time (s) 22
Application Rat lungs Solvable, proteinase K, nitric acid, and ammonium hydroxide were optimal at dissolving tissue Solvable emerged as the best solution because of its ability to remove background carbon, form compact CNT pellets, and minimize damage to the instrument (a) (b) (c) (d) (e) (f) (g) (h) Centirfuged rat lung tissue after treatment with the chemical digestion reagents: (a) Solvable, (b) hydroxide, (c) nitric, (d) sulfuric, (e) hydrochloric, (f) hydrofluoric, (g) peroxide, and (h) proteinase K. 23
Rat lung tissue treatment with Solvable only Some interference remains at higher temperatures where CNTs evolve Temperature ( C) 1 8 6 4 2 (a) Temperature FID.4.3.2.1 Carbon Mass (µg) 17 175 18 185 19 Time (s) 1 (b).4 Rat lung tissue treatment with Solvable and proteinase K Proteinase K successfully removed the remaining interfering carbon Temperature ( C) 8 6 4 2 Temperature FID.3.2.1 Carbon Mass (µg) 8 85 9 95 Time (s) 24
Extraction and quantification of CNTs in whole rat lungs Result: CNT Dose Alkali Treatment CNT 1 2 Transfer Enzymatic Treatment Analyze CNTs Collect CNTs A recovery of 93±15% of a 3.6 µg body burden deposited in individual whole rat lungs was achieved. 25
Graphene and in-situ reduction of graphene oxide Graphene oxide has a 1:1 O:C ratio How do we increase thermal strength?.25 1 9 Carbon Mass (µg).2.15.1 8 7 6 5 4 3 Temperature ( C).5 2 1 2 4 6 Time (s) Graphene Oxide Reduced Graphene Oxide Temperature 26
Application Composites Composite composition EPON 862 (Phenol-Formaldehyde Polymer Glycidyl Ether) EPIKURE W (diethylmethylbenzenediamine) Original CNT Graphene Treatment CNT or Graphene CNT or Graphene shavings pellet 27
Analysis of treated composite pellet 1.5 Temperature ( C) 8 6 4 Temperature Composite CNT Composite Graphene Composite.4.3.2 Carbon Mass (µg) 2.1 25 5 75 Time (s). 28
Going Forward Continue to develop extraction methods for other complex matrices (e.g., sediment) Develop alternative separation methods Finalize method for removing oxygen in-situ and examine effect on matrix carbon Is graphene and CNT separation possible? 29
Summary PTA is ideal for directly analyzing CNTs and graphene in simple matrices Extraction is necessary for complex matrices with embedded CNTs or a large amount of interfering carbon Extraction can also be used to concentrate samples with small CNT amount 3
Thanks! People Paul Westerhoff Rolf Halden Pierre Herckes Matt Fraser Andrea Clements Funding Support Science Foundation Arizona National Institutes of Health Semiconductor Research Corporation National Science Foundation Arizona State University (Ira A. Fulton, Paul and Elyse Johnson, GPSA) AZ Water 31