Skip to content

Thermal conduction of one-dimensional carbon nanomaterials and nanoarchitectures

MetadataDetails
Publication Date2018-03-17
JournalQUT ePrints (Queensland University of Technology)
AuthorsHaifei Zhan, Yuantong Gu
InstitutionsWestern Sydney University, Queensland University of Technology
AnalysisFull AI Review Included
  • This review maps nanoscale heat transport across one-dimensional carbon structures classified by carbon bonding: sp3 diamond nanowires and diamond nanothreads, sp2 carbon nanotubes and nanotube architectures, and sp1 carbyne.
  • Thermal conductivity is controlled by sample length, diameter, temperature, strain, junctions, defects, isotopic composition, functionalization, and inter-tube coupling.
  • Reported room-temperature conductivity spans more than five orders of magnitude, illustrating that carbon nanomaterials can support either efficient heat spreading or deliberate thermal insulation.
  • Representative values include approximately 3,000-3,500 W/m·K for carbon nanotubes, 1,750-5,800 W/m·K for single-walled CNT ropes, and about 5.4 W/m·K for ST(5,0) supernanotubes at 300 K.
  • Molecular dynamics using ReaxFF, AIREBO, Tersoff, and related potentials is the principal simulation method; reverse non-equilibrium molecular dynamics is widely used for finite-size conductivity calculations.
  • For practical thermal-management hardware, MPCVD SCD and PCD provide a manufacturable platform where high thermal conductivity, dimensional control, surface quality, and metal integration are required.
  • Material selection should distinguish axial nanoscale transport from device-level conduction across grain boundaries, interfaces, defects, contacts, and mounting layers.
ParameterValueUnitContext
Carbon-based thermal-conductivity spanMore than 5Orders of magnitudeAcross the reviewed 1D carbon nanomaterials and nanoarchitectures
CNT room-temperature thermal conductivity~3,000-3,500W/m·KRepresentative high-conductivity carbon nanotubes
SWCNT rope room-temperature conductivity1,750-5,800W/m·KExperimental measurement; strong axial transport
CNT array conductivity at density 0.81 g/cm3472W/m·KAxial conductivity at the lower reported array density
CNT array conductivity at density 1.39 g/cm3766W/m·KAxial conductivity at the higher reported array density
CNT inter-tube conductivity~0.35W/m·KNEMD result attributed to weak van der Waals coupling
Diamond nanowire Young’s modulus~688GPaSimulated <100>-oriented nanowire at 300 K
Diamond nanowire yield strength~63GPaSimulated <100>-oriented nanowire at 300 K
ST(5,0) finite-sample conductivity~5.4W/m·KReverse NEMD at 300 K
ST(5,0) extrapolated infinite-length conductivity~5.9W/m·KLinear extrapolation of inverse conductivity against inverse length
Supernanotube simulation range200-500KConductivity decreased continuously with increasing temperature
Carbyne conductivity~200-680W/m·KLength increased from 20 to 40 nm
Tensile-strain effect on carbyne~3% strain produced ~70% reduction—Small axial tension strongly suppressed thermal transport
Cumulene conductivity~83W/m·K50-carbon-atom metallic cumulene at 480 K, ReaxFF EMD
Cumulene electronic contribution1.11 × 10-3W/m·KNegligible relative to phonon transport
Polyyne conductivity~42W/m·KPristine chain at 500 K
Defective polyyne conductivity5.5W/m·K4% defective bonds at 500 K
Defective-bond fraction4%—Associated with a conductivity reduction from ~42 to 5.5 W/m·K
Cumulene-to-polyyne transition temperature499KTransition reported as complete within approximately 150 fs
High heating-rate threshold~2.5K/fsAbove this rate, defective-bond formation was observed
Defective carbon-bond length~1.45ÅLonger than a double bond but shorter than a single or triple bond
Cumulene HCACF long-time decay~15psLong-wavelength phonon relaxation component
Polyyne HCACF decay time~10psShorter than cumulene, explaining lower conductivity
Carbyne breaking force~10nNPredicted high stiffness and force required to break the chain
Graphdiyne nanotube-array thickness~15nmExperimental architecture noted in the review

Dominant Experimental and Modeling Variables

Section titled “Dominant Experimental and Modeling Variables”
ParameterObserved influenceEngineering significance
Sample lengthConductivity generally increases with length and approaches a finite limitDevice measurements must be corrected or compared at defined lengths
DiameterIncreasing diameter can reduce conductivity through lower group velocity and stronger Umklapp scatteringNanowire geometry affects thermal resistance
TemperatureConductivity commonly decreases as temperature rises because phonon-phonon scattering increasesPerformance should be specified at the intended operating temperature
Axial strainTension can substantially lower conductivity; compression may improve inter-tube transfer before bucklingStrain control is important in packaged devices
DefectsStructural defects create localized vibrations and strong phonon scatteringCrystal quality directly affects spreading efficiency
JunctionsCNT junctions can produce low-conductivity architectures suited to thermal barriersArchitecture can intentionally switch transport from high to low
DensityHigher CNT-array density increased conductivity from 472 to 766 W/m·KMorphology and packing influence macroscopic performance
Interfacial couplingWeak van der Waals contact produces very low cross-axis transportContact design and bonding can dominate over intrinsic material value
Isotopes and functionalizationAlter vibrational spectra and phonon scatteringEnable thermal-conductivity engineering
Grain boundariesScattering is generally stronger in PCD than in SCDSingle-crystal material is preferable when interface resistance is critical
  1. Classify structures by carbon hybridization

    • Group candidates as sp3 diamond nanowires and nanothreads, sp2 nanotubes and nanotube assemblies, or sp1 carbyne and related carbon chains.
    • Relate bonding topology to phonon dispersion, mechanical stiffness, and thermal transport.
  2. Construct or model nanoscale geometries

    • Generate pristine and defective diamond nanothreads, SWCNTs, MWCNTs, CNT fibers, CNT junctions, supernanotubes, and carbyne chains.
    • Vary structural parameters such as diameter, length, junction type, number of poly-benzene rings, and defect concentration.
  3. Select an interatomic potential

    • Use ReaxFF for structures involving bond formation, transition, or defects.
    • Use AIREBO for CNT and hydrocarbon-related interactions.
    • Use Tersoff-family or other validated potentials for diamond and other carbon systems.
    • Use DFT or first-principles methods where bond chemistry and electronic contributions require higher fidelity.
  4. Apply thermal boundary conditions

    • For reverse non-equilibrium molecular dynamics, divide the model into hot and cold reservoirs and establish a stationary temperature gradient.
    • Calculate heat flow, temperature slope, and conductivity from the imposed flux and resulting gradient.
    • Exclude or thermostat non-conductive boundary regions to prevent artificial reservoir effects.
  5. Account for finite size

    • Simulate multiple sample lengths and plot inverse conductivity against inverse length.
    • Extrapolate to infinite length to estimate intrinsic or saturated conductivity.
    • Treat the resulting values as model-dependent rather than universal material constants.
  6. Evaluate environmental variables

    • Sweep temperature, axial strain, diameter, density, defect fraction, and isotopic or functionalization state.
    • Compare pristine structures with defective variants to isolate phonon-scattering mechanisms.
  7. Analyze vibrational transport

    • Calculate vibrational density of states and frequency-resolved modes.
    • Evaluate heat-current autocorrelation functions to distinguish rapid high-frequency decay from slower long-wavelength relaxation.
    • Relate relaxation times to phonon group velocity and thermal resistance.
  8. Validate simulations with experiments

    • Compare predicted conductivity with measured values for CNT ropes, CNT arrays, and fabricated nanotubes.
    • Note that contact resistance, substrate effects, measurement method, sample density, and finite dimensions can create substantial variation.
  9. Translate results into a device architecture

    • Separate axial conduction from inter-tube or interface conduction.
    • Select high-conductivity SCD for compact heat spreaders and phonon-sensitive structures.
    • Consider PCD for larger-area structures where cost, scalability, and mechanical robustness are more important than minimal boundary scattering.
Research requirementRecommended 6CCVD materialRationale
Maximum phonon transmission with minimal grain-boundary resistanceOptical Grade SCDSingle-crystal continuity reduces grain-boundary phonon scattering and supports the highest device-level thermal performance
Large-area thermal spreadingOptical Grade PCDCombines diamond’s high thermal conductivity with scalable plates up to 125 mm
Conductive diamond electrodes with thermal functionalityHeavy Boron Doped PCDProvides electrical conductivity while retaining useful diamond thermal and chemical properties
Thin thermal paths or controlled interfacial spacingSCD from 0.1 µm to 500 µmEnables thin membranes and controlled insertion into layered thermal assemblies
Thick heat-spreader or substrate applicationsPCD or diamond substrate up to 10 mmSupports mechanically stable mounting and high heat flux
Moderate-conductivity or cost-optimized large componentsStandard PCDOffers scalable MPCVD diamond for components where the ultimate performance of SCD is unnecessary
Paper topicRelevance to 6CCVD capability
Diamond nanowires and nanothreadsThe paper identifies sp3 carbon as mechanically strong, chemically inert, and promising for nanoscale thermal management. 6CCVD SCD provides a high-purity bulk or thin-film basis for device-oriented extensions, although free-standing nanothread replication would require specialized downstream processing.
Length and thickness effects6CCVD can supply SCD and PCD from 0.1 µm to 500 µm, allowing the geometric dependence of thermal transport to be separated from substrate and contact effects.
Diameter and finite-size effectsDiamond membranes, patterned mesas, and laser-cut structures can be designed with controlled lateral dimensions for experimental validation.
Defect-controlled transportCrystal quality, growth conditions, surface preparation, and dopant concentration can be specified to tune scattering and conductivity.
CNT and supernanotube low conductivityThe low cross-axis or junction resistance described in the review highlights the value of 6CCVD PCD: numerous internal boundaries can provide useful thermal impedance when the design objective is suppression rather than spreading.
Carbyne and extreme length dependenceThe paper shows that a nominally high-conductivity nanoscale structure can lose performance sharply when length, defects, and strain change. SCD avoids the structural variability of free-standing atomic chains while retaining a well-defined crystalline thermal path.
  • Custom dimensions: PCD plates and wafers are available up to 125 mm, while SCD, PCD, and substrates can be produced over broad thickness ranges.
  • Surface finish: SCD can be polished to Ra &lt; 1 nm, supporting high-quality thermal interfaces and precise optical or electrical characterization.
  • PCD finish: Inch-size PCD can be polished to Ra &lt; 5 nm, providing a controlled surface for bonding and metallization.
  • Metalization: Internal Au, Pt, Pd, Ti, W, and Cu metallization options support heaters, thermometry, electrodes, solder or braze interfaces, and patterned thermal test structures.
  • Patterning: Custom laser cutting enables defined footprints, windows, bridges, and test coupons while reducing dependence on small, naturally occurring nanoscale structures.
  • Bonding architecture: Ti or W can be evaluated as adhesion or diffusion-barrier layers; Au, Pt, Pd, and Cu can be selected according to interface resistance, processing temperature, and attachment method.
  • Application variants: For a high-conductivity thermal spreader, SCD is preferred; for a larger, mechanically robust, lower-cost element, PCD is suitable; for a combined electrical and thermal function, boron-doped PCD is appropriate.
Device priorityRecommended design direction
Lowest possible thermal resistanceSCD, smooth surfaces, high-quality metal/diamond interface, minimized bond-line thickness, and limited defect density
Large-area heat spreadingPCD up to 125 mm, with thickness selected from thermal resistance and mechanical requirements
Simultaneous current delivery and heat removalHeavy Boron Doped PCD, with electrode geometry and metal stack optimized for current distribution
Controlled thermal resistanceAdjust diamond thickness, patterned geometry, grain structure, dopant level, and interface material
Reproducible thermal characterizationUse laser-cut coupons or patterned SCD membranes with defined length, width, thickness, and contact pads
High-temperature or harsh-environment operationSCD or PCD with a metal stack selected for the required temperature, atmosphere, adhesion, and chemical compatibility

For nanoscale thermal-management and high-power electronic-device research, begin with optical-grade SCD when the target is to isolate intrinsic diamond thermal transport or minimize grain-boundary resistance. Use PCD when the experiment requires larger device area, mechanical durability, or deliberate phonon scattering at internal boundaries. Heavy boron-doped PCD should be selected when the same component must also carry electrical current.

6CCVD’s in-house PhD team can assist with material selection, crystal-quality requirements, thickness, lateral geometry, surface preparation, and metal-stack design for similar high-power electronics, optoelectronics, photonics, diamond heat spreader, thermal-interface, and nanoscale transport projects.

For custom specifications or material consultation, visit 6ccvd.com or contact our engineering team directly.

View Original Abstract

This review summarizes the current studies of the thermal transport properties of one-dimensional (1D) carbon nanomaterials and nanoarchitectures. Considering different hybridization states of carbon, emphases are laid on a variety of 1D carbon nanomaterials, such as diamond nanothreads, penta-graphene nanotubes, supernanotubes, and carbyne. Based on experimental measurements and simulation/calculation results, we discuss the dependence of the thermal conductivity of these 1D carbon nanomaterials on a wide range of factors, including the size effect, temperature influence, strain effect, and others. This review provides an overall understanding of the thermal transport properties of 1D carbon nanomaterials and nanoarchitectures, which paves the way for effective thermal management at nanoscale.