DE-FE0031794
James M. Tour, Rice University, www.jmtour.com
Virtual Review: 20 October 2020
Proposal Title “Conversion of Domestic US Coal into Exceedingly High-Quality Graphene”
Main deliverable: Convert a US Coal Product Into 1 kg of Graphene Per Day
Project/Grant Period: 10/01/2019 through 9/30/2020
Flash Graphene
Discovered by
Duy Luong in
August 2018
2
Picture from videoblocks
Advantages of graphene
Graphene is:
• nontoxic and even used in several medical applications;
• is naturally occurring in the environment and agglomerates are the natural
mineral graphite;
• is a terminal natural sink for carbon since microbial decomposition is on
the order of hundreds of years, if at all, so it never again enters the carbon
cycle (i.e. graphite’s geological stability);
• can be used in composites of all types including plastics, which can be re-
flashed at end of life to make fresh graphene. So there are extreme energy
savings in lessening of the downstream composite hosts;
• At the current price of graphene being $60,000 to $200,000 per ton, there
is much room to capture markets;
• The electrical flashing cost is estimated to be $35 per metric ton of coal to
graphene conversion
Graphene Synthesis
LIG Applications
Electrochemical Exfoliation Chemical Exfoliation
Limitations for industrialization: Cons: 1. Highly defective product 2. Harmful by product
Cons: 1. Impurities 2. Quality control hard
1. Low product yield 3. Poor quality product 3. Complex synthesis
Top down 2. Environmental contamination Top down
Bottom up 3. Time-consuming Bottom up
4. Transfer & Integration
A new method is desired!
Epitaxial Growth Chemical Vapor Deposition
Cons: 1. Non-uniform product 2. High temperature required Cons: 1. High temperature required 2. Quality control hard
3. Time-consuming process 4. Low-yield 5. Hard transfer 3. Time-consuming process 4. Hard transfer 5. Harmful byproduct
Graphene Synthesis Approaches Review Kaichun Yang Thesis Defense 1/11/2019 2
DOI: 10.1038/s41570-017-0100
Flash Graphene Synthesis
LIG Applications
Carbon materials are converted efficiently into high quality
turbostratic graphene
Development of FG
Flash Unit Progression
FJH V0 sample
FJH Beta
FJH V0
FJH V1
FJH V1 sample
FJH V2.1
Flash Graphene
Apparatus
Flash Graphene Characterization
- Various carbon
LIG Applications
materials can be
converted into
graphene.
5 nm
- Graphene quality
is high.
- Graphene is
turbostratic
200 nm
which facilitates
dispersion in
composites
5 nm - $35/ton in
electrical energy
costs
100 nm
Time Temperature Dependence
LIG Applications
- Temperature needs to reach 3000 K for graphene conversion
- Pulse duration less than 50 ms for best graphene quality
- Cooling rate does not contribute to the quality of graphene
Turbostratic peaks in the
Raman spectrum of CB-FG.
IG/TS1 ~30. Lorentzian fit is
shown as a superimposed
smooth line. The R-squared
is 0.994 for TS1 and 0.99 for
TS2. These excellent fits
indicate the high quality of
the FG and the unmistakable
presence of these Raman
lines are attributable to
turbostratic graphene.
2D peak in the Raman spectrum of CB-FG. Left: best point in CB-FG, right: representative
point in CB-FG. Both peaks exhibit nearly a perfect Lorentzian line shape. The black dots are
the theoretical line shape. The R2 for the correlation is 0.999 for both peaks. This is indicative
of a fully conical Dirac cone at the K-point. The exceptionally large I2D/G is also indicative of
multilayer turbostratic graphene, as several researchers point to an increasing I2D/G.
Simulations
LIG Applications
Molecular dynamics (MD) simulations of structures with various characteristics (such as micro-porosity,
misalignment, and size of graphitic domains) kept at a given temperature range (1500 to 5000 K) for a
prolonged time (up to 5×10-9 s, with NVT thermostat). Sample structure after annealing at 3000 K for carbon
materials of various densities: a, 0.8 g·cm-3, sponge-like structure; b, 1.1 g·cm-3; c, 1.5 g·cm-3, high degree of
graphitization. d, Carbon black with 0.8 g.cm-3 density and large macro-porosity after prolonged (5×10-9 s)
annealing at 3600 K, polygonal fringes are apparent. e-f, Change of structural composition of materials during
annealing with different densities and temperatures. g. 1.5 g·cm-3 structure after annealing at 5000 K, initial
structure is the same as for c. Scale bars are 1.5 nm.
Flash Graphene Characterization
Flash graphene from “any” solid carbon source
LIG Applications
Carbon black, anthracite coal,
calcined coke, used coffee grounds,
charcoal, biochar, humic acid, keratin
5 nm
(human hair), lignin, sucrose, starch,
pine bark, olive oil soot, cabbage,
coconut, pistachio shells, potato
skins, rubber tires, polymers
including: polyethylene
terephthalate (PET or PETE), high- or
low-density polyethylene (HDPE,
LDPE), polyvinyl chloride (PVC),
polypropylene (PP), polyacrylonitrile
(PAN), or mixed plastics.
a) AB-stacked (Bernal) vs. b) turbostratic graphene
https://doi.org/10.1002/smll.200901173
Laboratory Scale-up FJH V2.1
FJH 5.0
16
Graphene can be used for so much
many
1 kg/day prototype will be
GMP-approved graphene
for medical purposes
www.universalmatter.com
FJH 5.0 Pill and Powder flashing
• Powder flashing • Pill flashing
- Flashing powder inside quartz tube - Flashing compressed pill in free form
• Pros • Pros
- Less volatile in feedstock lead to ease - Independent of a holder tube, desirable
of optimization for scale up system
• Cons • Cons
- Depends on a holder tube that is less - Need to find good binder with less
attractive for scale up volatiles
- Electrical contact need to be precise
FJH 5.0 flashing 1 g pill
• Metallurgical coke (MC) and raw green petroleum coke (GPC) are
mixed and pressed into pill. GPC acts as binder while MC acts as
conductor. Both are graphene source
• GPC and MC need to have the same size to make consistent pills
• Too large grain size yields pill inconsistency
• Pills are compressed with 15 mm die under 5 tons pressure
Sieve #20/30 pill Sieve #50/70 pill
19
FJH 5.0 pill optimization
MC/GPC #20-30 (0.6-0.84 #30-50 (0.3-0.6 #50-70 (0.21-0.30
mm) mm) mm)
50/50 Flashable Flashable Flashable
30/70 Inconsistent Inconsistent Inconsistent
70/30 Pill break Flashable Flashable
9 pills compositions
Voltage Frequency (Hz)/ duty cycle (%), ON/OFF (ms)
200/10 200/20 200/50 1000/10 1000/20 1000/50
0.5/4.5 1/4 2.5/2.5 0.1/0.9 0.2/0.8 0.5/0.5
160 20-30% 30-40% Lost form 10-20% 20-40% 20-40%
200 30-40% Lost form Lost form 30-40% 30-40% Lost form
12 basic flashing parameters (50/50 #30-50 sample) 20
FJH 5.0 flashing 1 g pill
• Flashing jig for pill
21
Variable Frequency Drive flashing
• Using pulse width modulation, we can turn DC flashing into
AC flashing
• The result is homogeneous and controllable flash
• Raman shows graphene yield > 80%
3500
1600
3000
1400
2500 1200
1000
Current (A)
Current (A)
2000
800
1500
600
1000
400
500
200
0
0
-500 -200
20 20.2 20.4 20.6 20.8 21 13.8 14 14.2 14.4 14.6 14.8 15 15.2 15.4 15.6 15.8 16
Time (s)
Time
Time(s)
Simple DC flash brings the current up to VFD flash allows controllable current
more than 3000 A and result in a less and stop the chain reaction.
controllable reaction.
22
Variable Frequency Drive flashing
• Using variable duty cycle in the same flash, we can
further optimize the flash. Average Raman
1600
• Freestanding flash is demonstrated for the first time 1400
with multiple VFD flashing 1200
• Process yield of 40% and 100% graphene yield 1000
Intensity
800
2500
600
400
2000 1600
200
1400 0
Current (A)
1500
-200
1200 500 1000 1500 2000 2500 3000 3500
1000 1000 Wavenumber (cm-1)
Current (A)
800
500
600
0 400
14.5 15 15.5 16 16.5
200
Time (s)
time (s)
Single VFD result in triangle current profile 0
11 11.2 11.4 11.6 11.8 12 12.2 12.4 12.6 12.8
Time (s)
Multiple VFD results in prolonged shape current profile
23
FJH 5.0 powder flashing
• 5 g batch
• Homogeneous grain size powder provides even flash
• High product yield (flashed material/feedstock) of 90%
• ~100% graphene yield (graphene from flashed material, determined by Raman)
• With VFD flashing, no pretreatment is needed while the graphene yield increases with faster flashing. In 2 h we have made 30 g
of flash graphene (0.25 g/min).
• Further manual flashing optimization can result in 1 g/min (1.4 kg/day)
2500
Average Raman
2000 spectrum shows
high quality of
1500 graphene with low
Intensity
D peak and high 2D
1000
500
0
500 1000 1500 2000 2500 3000 3500
Wavenumber (cm-1)
24
FJH 5.0 powder flashing
• 0.2-1 µm in size
• Turbostratic structure seen by HRTEM and FFT.
25
Powder flashing automation
• Building automation using 3D printed parts
• Aiming for 1 batch/min (5 g/min, 7.2 kg/24 h).
• 5.7 g of metallurgical coke (MC) per batch to
provide > 5.0 g MC flash graphene per batch
• Voltage: 350-400 V, VFD flashing at 1000 Hz, 20%
duty cycle
• Process yield: > 90%, graphene yield: ~100%.
26
VFD PID control
• Multiple VFD for current control
• Control loop base on current and temperature
27
New Raman system
• Install new Raman system
• Capable of autofocus and mapping
• 121 Raman spectra in 10 min, 10x faster
28
Worldwide Consumption of Materials
per year Embodied
Million tons* Energy**
Cement: 3,000
4,000 >26B GJ
Concrete: 40,000
30,000
Timber: 3,200
Timber for construction: < 1,000
All metals: 1,000
Steel: 940
18B GJ
Steel for construction: < 400
Paper and cardboard: 300
Plastics: 120
Aluminium: 32
Concrete energy:
>600 Nuclear Power Plants
>1800 Hoover Dams
Concrete production produced 5-8% of CO2 emissions yearly.
https://www.statista.com/statistics/219343/cement-production-worldwide/
Summary
• Task 3.4: Optimize FJH 5.0 for 10 g flash unit
Due 1 Dec 2020, replaced with VFD flashing of 6
g batches
• Task 3.5: Design of scale-up reactor for 100 g
flash
Due 1 March 2021, replaced with automation
• Task 3.6: Build FJH 6.0 for 100 g flash unit
Due 1 June 2021, on schedule
30
= Flash workers
Thanks to the DOE for support. DE-FE0031794
And Jason Hissam for patience during 2.5 months of lab closure
