Addressing the Trade-Off between Crystallinity and Yield in Single-Walled Carbon Nanotube Forest Synthesis Using Machine Learning.

Synthetic trade-offs exist in the synthesis of single-walled carbon nanotube (SWCNT) forests, as growing certain desired properties can often come at the expense of other desirable characteristics such as the case of crystallinity and growth efficiency. Simultaneously achieving mutually exclusive properties in the growth of SWCNT forests is a significant accomplishment, as it requires overcoming these trade-offs and balancing competing mechanisms. To address this, we trained a machine-learning regression model with a set of 585 "real" experimental synthesis data, which were taken using an automatic synthesis reactor. Subsequently, 16000 exploratory "virtual" experiments were performed by our trained model to examine potential routes toward addressing the current crystallinity-height trade-off limitation, and suggestions on growth conditions were predicted. Importantly, additional validation using "real" experimental syntheses showed good agreement with the predictions as well as a 48% increase in growth efficiency while maintaining the high crystallinity (G/D-ratio). This highlighted the effectiveness and accuracy of the predictive capability of our machine-learning model, which achieved improved results in less than 50 validation tests. Furthermore, the trained model revealed the surprising importance of the nature of the carbon feedstock, particularly the reactivity and concentration, as a route for overcoming the trade-off between the SWCNT crystallinity and growth efficiency. These results of the high-efficiency synthesis of highly crystalline SWCNT forests represent a significant advance in overcoming synthetic trade-off barriers for complex multivariable systems.

Enhanced activity for reduction of 4-nitrophenol of Ni/single-walled carbon nanotube prepared by super-growth method.

In this study, we synthesised the Ni/single-walled carbon nanotube prepared by the super-growth method (SG-SWCNTs). In this approach, the Ni nanoparticles were immobilised by an impregnation method using the SG-SWCNTs with high specific surface areas (1144 m2g-1). The scanning electron microscopy images confirmed that the SG-SWCNTs exhibit the fibriform morphology corresponding to the carbon nanotubes. In addition, component analysis of the obtained samples clarified that the Ni nanoparticles were immobilised on the surface of the SG-SWCNTs. Next, we evaluated the activity for the reduction of 4-nitoropenol in the presence of the Ni/SG-SWCNTs. Additionally, the Ni/graphene, which was obtained by the same synthetic method, was utilised in this reaction. The rate of reaction activity of the Ni/SG-SWCNTs finished faster than that of the Ni/GPs. From this result, the pseudo-first-order kinetic rate constantkfor the Ni/SG-SWCNTs and the Ni/GPs was calculated respectively at 0.083 and 0.070 min-1, indicating that the Ni/SG-SWCNTs exhibits higher activity.

The relationship between the growth rate and the lifetime in carbon nanotube synthesis.

We report an inverse relationship between the carbon nanotube (CNT) growth rate and the catalyst lifetime by investigating the dependence of growth kinetics for ∼330 CNT forests on the carbon feedstock, carbon concentration, and growth temperature. We found that the increased growth temperature led to increased CNT growth rate and shortened catalyst lifetime for all carbon feedstocks, following an inverse relationship of a fairly constant maximum height. For the increased carbon concentration, the carbon feedstocks fell into two groups where ethylene/butane showed an increased/decreased growth rate and a decreased/increased lifetime indicating different rate-limiting growth processes. In addition, this inverse relationship held true for different types of CNTs synthesized by various chemical vapor deposition techniques and continuously spanned a 1000-times range in both the growth rate and catalyst lifetime, indicating the generality and fundamental nature of this behavior originating from the growth mechanism of CNTs itself. These results suggest that it would be fundamentally difficult to achieve a fast growth with a long lifetime.

Diameter control of single-walled carbon nanotube forests from 1.3-3.0†nm by arc plasma deposition.

We present a method to both precisely and continuously control the average diameter of single-walled carbon nanotubes in a forest ranging from 1.3 to 3.0 nm with ~1 Šresolution. The diameter control of the forest was achieved through tuning of the catalyst state (size, density, and composition) using arc plasma deposition of nanoparticles. This 1.7 nm control range and 1 Šprecision exceed the highest reports to date.

The infinite possible growth ambients that support single-wall carbon nanotube forest growth.

We report the virtually infinite possible carbon feedstocks which support the highly efficient growth of single-wall carbon nanotubes (SWCNTs) using on the water-assisted chemical vapor deposition method. Our results demonstrate that diverse varieties of carbon feedstocks, in the form of hydrocarbons, spanning saturated rings (e.g. trans-deca-hydronaphthalene), saturated chains (e.g. propane), unsaturated rings (e.g. dicyclopentadiene), and unsaturated chains (e.g. ethylene) could be used as a carbon feedstocks with SWCNT forests with heights exceeding 100 ums. Further, we found that all the resultant SWCNTs possessed similar average diameter indicating that the diameter was mainly determined by the catalyst rather than the carbon feedstock within this synthetic system. A demonstration of the generality was the synthesis of a carbon nanotube forest from a highly unorthodox combination of gases where trans-decahydronaphthalene acted as the carbon feedstock and benzaldehyde acted as the growth enhancer.

Absence of an ideal single-walled carbon nanotube forest structure for thermal and electrical conductivities.

We report the fundamental dependence of thermal diffusivity and electrical conductance on the diameter and defect level for vertically aligned single-walled carbon nanotube (SWCNT) forests. By synthesizing a series of SWCNT forests with continuous control of the diameter and defect level over a wide range while holding all other structures fixed, we found an inverse and mutually exclusive relationship between the thermal diffusivity and the electrical conductance. This relationship was explained by the differences in the fundamental mechanisms governing each property and the optimum required structures. We concluded that high thermal diffusivity and electrical conductance would be extremely difficult to simultaneously achieve by a single SWCNT forest structure within current chemical vapor deposition synthetic technology, and the "ideal" SWCNT forest structure would differ depending on application.

Unexpectedly high yield carbon nanotube synthesis from low-activity carbon feedstocks at high concentrations.

We report a new direction for highly efficient carbon nanotube (CNT) synthesis where, in place of conventional highly reactive carbon feedstocks at low concentrations, highly stable carbon feedstocks at high concentrations were shown to produce superior yields. We found that a saturated hydrocarbon that is considered to possess a low reactivity, delivered at high concentrations, could achieve an extremely high growth yield (2.5 times that when using ethylene). This result stems from the unique behavior where the CNT yield linearly increased with carbon concentration, in contrast to more reactive carbon feedstocks, where the yield peaks. We propose that the mechanisms for the growth kinetics for high- and low-reactivity carbon feedstocks are fundamentally different, where the latter benefits from a longer catalyst lifetime because of a relatively low production rate of carbon impurities.

Mutual exclusivity in the synthesis of high crystallinity and high yield single-walled carbon nanotubes.

We report the mutually exclusive relationship between carbon nanotube (CNT) yield and crystallinity. Growth conditions were optimized for CNT growth yield and crystallinity through sequential tuning of three input variables: growth enhancer level, growth temperature, and carbon feedstock level. This optimization revealed that, regardless of the variety of carbon feedstock and growth enhancer, the optimum conditions for yield and crystallinity differed significantly with yield/crystallinity, preferring lower/higher growth temperatures and higher/lower carbon feedstock levels. This mutual exclusivity stemmed from the inherent limiting mechanisms for each property.