Author: admin

Apple has been hit with an AI lawsuit in a U.S. federal court, where authors allege the company used copyrighted books without consent or compensation to train its artificial intelligence systems.

The proposed class action was filed on Friday in the federal court for Northern California by authors Grady Hendrix and Jennifer Roberson. The lawsuit claims that Apple copied protected works without seeking permission, credit or compensation from the writers.

“Apple has not attempted to pay these authors for their contributions to this potentially lucrative venture,” the filing stated. The lawsuit further alleges that Apple relied on a known body of pirated books to train its large language models under the “OpenELM” project.

Hendrix, who resides in New York, and Roberson, based in Arizona, said their works were included in the pirated dataset. Apple and legal representatives for the plaintiffs did not immediately respond to media requests for comment.

The Apple AI lawsuit is the latest in a series of legal challenges targeting major technology firms accused of using copyrighted material to train artificial intelligence systems. Earlier on Friday, AI startup Anthropic disclosed in a California court filing that it had agreed to pay $1.5 billion to settle a class action brought by authors who accused the company of using their works without approval to train its chatbot Claude. Lawyers for the plaintiffs described it as the largest publicly reported copyright recovery in history, though Anthropic did not admit liability.

In June, Microsoft was also sued by a group of authors alleging their books were used without consent to train its Megatron AI model. Meta Platforms and OpenAI, backed by Microsoft, have faced similar claims over the alleged misuse of copyrighted content.

The lawsuit against Apple adds to the growing wave of intellectual property disputes surrounding artificial intelligence, as authors, publishers and content creators seek stronger protections in the rapidly evolving AI landscape.

Introduction

Chemical vapour deposition (CVD) is one of the most extensively studied methods for synthesising nanoparticles.^1,2 CVD is an exceptionally versatile method that enables the formation of a wide range of nanoparticles, including carbon nanotubes, graphene, hexagonal boron nitride, MXenes, and many more 1D and 2D materials.^3–6 Its advantages, such as simplicity, scalability, and economic efficiency, show great promise for potential industrial manufacturing of synthesized materials. Since its discovery over 50 years ago, many types of CVD processes have been distinguished, including water-assisted CVD,^7,8 plasma-enhanced CVD (PE-CVD),^9,10 metal-organic CVD (MOCVD)¹¹ or floating catalyst CVD (FC-CVD).^12,13 The latter one, FC-CVD, has gained substantial interest due to the unique opportunities it offers in the synthesis of carbon nanotubes,^14–17 particularly enabling a continuous one-step synthesis of CNT macroarchitectures such as CNT fibres and films, as well as CNT arrays.^4,18,19 Furthermore, FC-CVD offers the possibility of controlling the nanostructure of CNTs. The process can be optimised to synthesise single-walled carbon nanotubes (SWCNT),^17,20,21 double-walled carbon nanotubes (DWCNT)²² and multi-walled carbon nanotubes (MWCNT),²³ as well as provide control over the length, chirality or doping of the nanotubes.^20,24,25 Understanding that CVD processes are capable of synthesising a wide range of nanoparticles, it is reasonable to anticipate that FC-CVD might offer capabilities beyond merely synthesising traditional carbon nanotubes, through adjustments in process parameters and feedstocks. In the quest for precise synthesis of CNTs, a plethora of process variables have been explored, including temperature settings, types and flow rates of gases, reactor configurations, carbon sources, sulphur compounds, and, lastly, catalysts.^17,24,25 Regarding the latter parameter, one of the earliest studies of FC-CVD process by Sen et al focused on the use of metallocenes.²⁶ The metallocenes tested in this research paper included ferrocene, cobaltocene and nickelocene. All these metallocenes were further studied separately as well as in mixtures.^27,28 The key studies relied on using an iron catalyst derived from ferrocene^{20,23,29–31} due to much better yield and quality of CNTs, improved growth rate and scalability of the process.³² Nonetheless, it has been shown that the use of cobaltocene may lead to the formation of interesting structures such as cup-stacked carbon nanotubes²⁷ as well as improved catalytic activity in the case of cobaltocene and nickelocene mixtures with ferrocene.^28,33 This indicates that the use of other nonferrous metallocenes may alter carbon nanostructure synthesis routes, potentially yielding novel nanomorphologies. To explore this opportunity, in this work, we conducted a series of experiments using six compounds from the metallocene family: Ferrocene, Cobaltocene, Ruthenocene, Vanadocene, Manganocene, and Magnesocene as catalyst sources in FC-CVD syntheses of carbon nanomaterials. These experiments revealed the emergence of novel structures, diverging from conventional carbon nanotubes, some featuring distinctive shapes. In particular, the unique dendritic morphologies obtained with manganocene suggest potential applications in areas such as energy storage, sensing, or printed electronics, where high surface area and oxygen functionality are advantageous.

Results and Discussion

The first experiment aimed to obtain the reference data for further studies presented in this work. The experiment followed the procedure identified in the literature to be the most efficient, yielding carbon nanotube arrays of the highest purity.^34–36 The CNT arrays were synthesised using a solution of 5 wt.% of ferrocene in toluene. The preheater temperature was set to 170°C. The evaporating feedstock has been injected into the hot zone of the reactor and transported with the argon flow set at 1 L/min. The synthesis has been performed at 760°C. SEM images and Raman spectrum of synthesised CNT structure are presented in Figure 1.

Figure 1 Scanning electron microscopy (a) and Raman spectrum (b) of carbon nanotube array structure obtained in the dedicated process with the ferrocene as the catalyst.

As is shown in Figure 1 the FC-CVD method allowed obtaining a regular array structure of carbon nanotubes. Individual carbon nanotubes had typical diameters in the range of several tens of nanometers. The height of the arrays was about one millimetre. Such a result ranks very well among the CNT array structures described in the literature^34–38 and proves that the applied parameters of the process were well-adjusted.

The presented Raman spectrum shows all features characteristic of Raman spectroscopy of CNT structures. Radial breathing mode (RBM) is visible at low frequencies, suggesting that the synthesised material includes single-wall nanotubes (SWNT). Around 1330 cm⁻¹, a D-band peak appears, which is typically observed in all graphitic structures that contain some impurities or defects that break the sp² carbon structure symmetry.^39–41 Next, at the frequency of ~1580 cm⁻¹ G-band, originating from carbon atom vibrations is visible. The overtone of the D-band, known as the 2D-band, is also visible on the analysed spectrum at ~2650 cm⁻¹. All the peaks are sharp and well-located, which proves the good quality and purity of the material.

In the following syntheses, the ferrocene catalyst precursor has been replaced by other metallocenes – Cobaltocene, Ruthenocene, Vanadocene, Manganocene and Magnesocene. The process conditions which were found to yield the best CNT arrays for ferrocene-based synthesis were assumed to be the starting point for other metallocene syntheses. However, some procedure parameters had to be adjusted to accommodate the melting points and solubility of the new catalysts being tested. These compounds share a closely similar structure. All of them are bis(cyclopentadienyl)metal structures, however, they exhibit some differences, for example, having varying melting points with values: 172.5°C, 173°C, 194°C, 165°C, 173°C, 176°C for Ferrocene, Cobaltocene, Ruthenocene, Vanadocene, Manganocene and Magnesocene, respectively.

For all catalysts, a feedstock of 5 wt.% metallocene in toluene was prepared. During the preparation of the feedstock, different solubilities of individual metallocenes were observed. This resulted in an extended sonication-assisted dissolution time, up to an hour, in the case of the Vanadocene and Manganocene.

For all the tested metallocenes, experiments were first carried out with the process parameters used for ferrocene (preheater temperature of 170°C, oven temperature – 760°C). Next, the temperatures were modified in the range of 650–850°C for the furnace, and 150–200°C at the preheater and product deposition on quartz glass was investigated. For most of the catalysts tested, modifications of process temperatures did not improve the deposition of the products. However, the significant difference in the melting point of ruthenocene compared to ferrocene necessitated an increase of the temperature of the preheater to 200°C and the temperature of the actual pyrolysis to 850°C.

All deposits obtained in the syntheses were subject to scanning electron microscopy (SEM) and Raman spectroscopy analysis. SEM images of the obtained nano and micromaterials are presented in Figure 2.

Figure 2 Scanning electron microscopy images of carbon structures synthesised with the usage of (a and b) cobaltocene, (c and d) magnesocene, (e and f) vanadocene, (g and h) ruthenocene, (i and j) manganocene as a catalyst. For each catalyst, images are presented at 10,000× (left) and 25,000× (right) magnification to allow consistent comparison of morphological features.

As shown in Figure 2a and b, the use of cobaltocene as a reaction catalyst formed nanotube structures. However, the nanotubes are not arranged in a regular array structure, as is the case with iron catalysts.

In the case of the Magnesocene, individual nanotube structures can be seen at higher magnification (Figure 2d). However, their number is negligible and their length is much shorter than for the Cobaltocene. In the case of the catalyst in the form of both a Vanadium compound (Figure 2e and f) and a Ruthenium compound (Figure 2g and h), no nanotube structures are visible. The structures visible in the obtained images are rather similar to graphite forms decorated with metal catalysts. The analysis of SEM images seems to indicate that the most interesting material is the product of the manganocene-catalysed reaction (Figure 2i and j). Dendritic structures resembling fern leaves were obtained.

The observed variations in carbon nanostructure morphology can be attributed to the intrinsic properties of the metallocene catalysts used. Thermal decomposition temperatures play a crucial role; metallocenes with lower decomposition points activate earlier, leading to different catalyst particle sizes and distributions. Moreover, the strength of metal-carbon interactions influences the formation of metal carbides, which serve as active sites for carbon growth. For instance, nickel’s strong affinity for carbon facilitates the formation of stable carbides, promoting the growth of well-structured carbon nanotubes. In contrast, metals with weaker carbon interactions may lead to amorphous carbon structures. Additionally, the electronic configuration and oxidation state of the metal center affect catalytic activity and selectivity, further influencing the morphology of the synthesized carbon materials.

All obtained materials were also analysed by Raman spectroscopy for chemical and structural identification (Figure 3a–e). Due to the unusual and intriguing structure of the manganocene-catalysed material, it was additionally subjected to an Energy Dispersive Spectroscopy (EDS) analysis to characterise the elemental composition (Figure 3f).

Figure 3 Raman spectra of carbon structures synthesised with the usage of (a) cobaltocene, (b) magnesocene, (c) vanadocene, (d) ruthenocene, (e) manganocene as a catalyst. Energy Dispersive Spectroscopy analysis of the structure synthesised with manganocene as a catalyst (f).

The Raman spectra presented in (Figure 3a–e) seem to lead to conclusions consistent with the analysis of SEM images. For the structures obtained for the cobaltocene (Figure 3a), magnesocene (Figure 3b), ruthenocene (Figure 3c) and vanadocene (Figure 3d) catalysis, the Raman spectra show the D and G bands characteristic of nanocarbon structures, with wavenumbers of around 1300 and around 1580 cm⁻¹, respectively. The bands in these positions are characteristic of sp² hybridized materials. Thus, both nanotube structures and graphene/graphite can be present in the samples. Therefore, to analyse the composition, it is necessary to correlate the results of Raman spectroscopy with SEM images. In the case of cobalt (Figure 3a) and magnesium (Figure 3b) based catalysts, the images clearly show the presence of carbon nanotubes. The G-band to the D-band intensity ratio is often mentioned as a useful tool for determining the quality of produced carbon nanotubes. The D-band is associated with defects within the graphic lattice or even with amorphous carbons. The better the purity and quality of the nanotubes, the higher the value of the G:D ratio is observed.⁴² Therefore, both the cobaltocene and magnesocene carbon nanotubes contain numerous defects, which is also consistent with the results of the SEM analysis.

Furthermore, taking into account both the SEM imaging results and the presence of the typical G and D bands in Raman spectroscopy for materials catalysed by ruthenocene (Figure 3c) and vanadocene (Figure 3d), it can be concluded that graphite structures were obtained, probably functionalised to some extent by the presence of ruthenium and vanadium elements, respectively.

Even though the presence, shape and location of the characteristic bands provide a large amount of information about the carbon material, there is a large discrepancy in the literature regarding the Raman spectra of various graphite systems, eg relatively newly produced graphitised structures reduced from oxides, wrinkled or functionalised sp²-bonded carbon materials.^43–46 For graphene, a strong share of the 2D band is usually observed, which is not observed for the materials tested here. Yet, it is not difficult to find reports in which the 2D band is very broadened and not very intense, as well as in which the 2D band is omitted at all.^46–49 In general, Raman spectra of graphitised structures depend not only on the number of layers or the number of graphitisation disturbances but also largely on their functionalisation, production method or substrate.

Among all tested materials the most intriguing and thought-provoking is the one catalysed by the manganocene (Figure 3e). With the aim to better analyse the surprising SEM images and the diversity of bands in Raman spectroscopy, this material was also subjected to elemental composition analysis using EDS (Figure 3f). As shown by the result of this test, the compound synthesized as a result of the Manganocene catalysed reaction consists of carbon, manganese and oxygen. This most likely indicates that nanocarbon structures, built on the manganese catalyst were obtained, which are further functionalized with oxygen from the air. This information allows a better understanding of the plurality of bands in the Raman spectrum.

A number of bands can be observed in the Raman spectrum of this material, some of which can be sorted into groups. The first visible signal is around 131 cm⁻¹, the intense band in this position for carbon structures is often associated with the presence of single-walled nanotubes. Between 1069 cm⁻¹ and 1371 cm⁻¹, a group of three bands is visible, and these are the most intense peaks of the analysed spectrum. The location of these bands, namely in the vicinity of the D band, may indicate that we are dealing with a highly disturbed structure of nanotubes. For wavenumber 1722 cm⁻¹, a single band is visible, followed by another triplet of bands at positions between 2013 cm⁻¹ and 2373 cm⁻¹. The last visible band appears at the wavenumber of 2903 cm⁻¹. The analysis of such arranged strands is not easy. Both the SEM images, which show the extremely complex architecture of the material, and the EDS results, which show a small number of constituent elements, should be considered. Bands in the range 1650–1750 cm⁻¹ indicate the presence of carbon-oxygen double bonds and the bands in the range 2100–2300 cm⁻¹ indicate the presence of carbon-carbon triple bonds. The last band, on the other hand, indicates carbon-hydrogen vibrations.⁵⁰ The structure obtained is probably a nanocarbon structure with a strong disturbance of the graphite structure, for example, due to the complicated structure and the related series of carbon bonds. In addition, this structure is most likely functionalized with oxygen, which presumably happened due to the oxidation of the carbon structures by the manganese catalyst.

While this study focused on the synthesis and structural analysis of carbon nanomaterials derived from various metallocene catalysts, the unique features of the manganese-catalysed dendritic structures—such as high surface complexity, presence of oxygen functionalities (as suggested by EDS and Raman), and hierarchical morphology—make them promising candidates for further functional evaluation.

In particular, such materials could be explored in printed electronics, as novel functional fillers or electrodes in conductive inks and pastes,^51,52 or in electrochemical sensing platforms,^53,54 where surface functional groups can enhance molecular binding. Moreover, their complex surface topology and redox-active components suggest potential in energy storage devices,⁵⁵ including supercapacitors⁵⁶ and battery electrodes.⁵⁷ These possibilities position the synthesised materials as intriguing subjects for future application-driven studies.

Conclusions

Unique properties of nanoscopic carbon materials lead to extensive works of both the new materials among this family and other synthesis possibilities of known materials. To fulfil both of those trends we presented research on the capabilities of CNT arrays-dedicated furnace, on the synthesis of either nanotube with better alignment or new carbon nanostructures. The use of metallocenes as catalysts was associated with the proven effectiveness of ferrocene in the synthesis of good quality vertically aligned carbon nanotubes carpets. Conducted tests showed that catalysis by cobaltocene and magnesocene allows obtaining the structure of nanotubes. However, these are not regularly arranged parallel structures and graphitization is strongly disturbed. The defects are stronger for the catalysis with the magnesium compound. The materials synthesized by the vanadocene and ruthenocene catalysis did not show the structure of the nanotubes. However, the Raman spectra of these materials reveal the G – and D – bands characteristic of the graphite structures. The observations of the scanning electron microscopy images allow confirming the presence of graphite structures. The manganocene-catalysed reaction allows for obtaining very interesting structures resembling ferns. Based on the conducted material analysis we were able to characterize the material as oxidized and functionalized carbon structures. The manganocene-catalysed reaction allows for obtaining very interesting structures resembling ferns. Based on the conducted material analysis we were able to characterize the material as oxidized and functionalized carbon structures.

The research conducted allowed us to obtain interesting materials. It was shown that the growth of nanotube structures does not occur for the entire family of metallocenes and that intriguing and unusual structures can be obtained. These findings highlight the potential of nonferrous metallocenes to diversify the morphology of carbon nanomaterials beyond traditional CNTs. Considering the unique features and potential applications of these nanostructures, particularly those derived from manganocene, it would be highly valuable to pursue further studies. Future experimental research should focus on optimising synthesis conditions, assessing reproducibility, addressing the scalability of material manufacturing, as well as evaluating the functional performance of these novel nanostructures, including the testing of electrical, thermal and mechanical properties, and chemical reactivity. Full understanding of the mechanisms responsible for the synthesis of novel nanomaterials could be gained through the simultaneous atomistic modelling of the synthesis reactions.

Experimental Section

Materials: In all experiments, toluene was used as a carbon substrate. The tested catalysts included Ferrocene, Cobaltocene, Ruthenocene, Vanadocene, Manganocene, and Magnesocene. Toluene and Ferrocene were purchased from Sigma Aldrich, UK. Other metallocenes were purchased from Strem Chemicals, Inc., UK.

Synthesis

Carbon nanomaterials were synthesized using the previously described chemical vapour deposition process, tailored specifically for the synthesis of carbon nanotube arrays.⁵⁸ Briefly, the setup comprised a horizontal tube furnace capable of reaching temperatures up to 900°C, a liquid feedstock injector heated to 200°C, a source of argon gas, and an exhaust filtering system. During a standard CNT synthesis process, the liquid feedstock, comprised of a carbon source and a catalyst source, is heated and then injected into the reactor’s main zone with a flow of argon. Within the reactor’s hot zone, the feedstock undergoes pyrolysis. Iron catalysts deposit on the walls of the quartz tube, initiating the growth of CNTs in the form of arrays. All syntheses were performed in triplicate to confirm reproducibility. Representative results are shown.

Solutions Preparation

The metallocene solutions in toluene were prepared by weighing the desired amount of metallocene and toluene and their homogenization using a VCX 750 ultrasonic homogeniser until visible dissolution of metallocene in toluene. 10 minutes each for Ferrocene, Cobaltocene, Ruthenocene, Magnesocene. 60 minutes each for, Vanadocene, and Manganocene.

Characterization

The Raman analysis was made on the Horiba LabRam 300 spectrometer with a 17 mW, 633 nm red laser. The analysis software was Lab Spec 5. Scanning Electron Microscopy of the reference CNT array was carried out using a Carl Zeiss Auriga 60 high-resolution scanning electron microscope with an accelerating voltage of 8 kV. Structural investigations were combined with advanced energy-dispersive spectroscopy analysis.

Funding

This research was funded by Warsaw University of Technology IDUB, POB Materials Technologies – 3 ADVANCED grant no1820/359/Z01/POB5/2021.

Disclosure

The authors report no conflicts of interest in this work.

References

1. Allaedini G, Tasirin SM, Aminayi P. Synthesis of CNTs via Chemical Vapor Deposition of Carbon Dioxide as a Carbon Source in the Presence of NiMgO. J Alloys Compd. 2015;647:809–814. doi:10.1016/j.jallcom.2015.06.012

2. Kumar M, Ando Y. Chemical Vapor Deposition of Carbon Nanotubes: a Review on Growth Mechanism and Mass Production. J Nanosci Nanotechnol. 2010;10(6):3739–3758. doi:10.1166/jnn.2010.2939

3. Hussein FH, Abdulrazzak FH. Synthesis of Carbon Nanotubes by Chemical Vapor Deposition. In: Nanomaterials: Biomedical, Environmental, and Engineering Applications. John Wiley & Sons, Ltd. 2018:105–132. doi:10.1002/9781119370383.ch4

4. Uchida Y, Kawahara K, Fukamachi S, Ago H. Chemical Vapor Deposition Growth of Uniform Multilayer Hexagonal Boron Nitride Driven by Structural Transformation of a Metal Thin Film. ACS Appl Electron Mater. 2020;2(10):3270–3278. doi:10.1021/acsaelm.0c00601

5. Fan Y, Li L, Zhang Y, Zhang X, Geng D, Hu W. Recent Advances in Growth of Transition Metal Carbides and Nitrides (MXenes) Crystals. Adv Funct Mater. 2022;32(16):2111357. doi:10.1002/adfm.202111357

6. Bhowmik S, Govind Rajan A. Chemical Vapor Deposition of 2D Materials: a Review of Modeling, Simulation, and Machine Learning Studies. iScience. 2022;25(3):103832. doi:10.1016/j.isci.2022.103832

7. Hata K, Futaba DN, Mizuno K, Namai T, Yumura M, Iijima S. Water-Assisted Highly Efficient Synthesis of Impurity-Free Single-Walled Carbon Nanotubes. Science. 2004;306(5700):1362–1364. doi:10.1126/science.1104962

8. Patole SP, Alegaonkar PS, Lee H-C, Yoo J-B. Optimization of Water Assisted Chemical Vapor Deposition Parameters for Super Growth of Carbon Nanotubes. Carbon. 2008;46(14):1987–1993. doi:10.1016/j.carbon.2008.08.009

9. Wu Y, Qiao P, Chong T, Shen Z. Carbon Nanowalls Grown by Microwave Plasma Enhanced Chemical Vapor Deposition. Adv Mater. 2002;14(1):64–67. doi:10.1002/1521-4095(20020104)14:1<64::AID-ADMA64>3.0.CO;2-G

10. Li Y, Mann D, Rolandi M, et al. Preferential Growth of Semiconducting Single-Walled Carbon Nanotubes by a Plasma Enhanced CVD Method. Nano Lett. 2004;4(2):317–321. doi:10.1021/nl035097c

11. Lee DH, Sim Y, Wang J, Kwon S-Y. Metal–Organic Chemical Vapor Deposition of 2D van Der Waals Materials—The Challenges and the Extensive Future Opportunities. APL Mater. 2020;8(3):030901. doi:10.1063/1.5142601

12. Zhou Z, Ci L, Song L, et al. Producing Cleaner Double-Walled Carbon Nanotubes in a Floating Catalyst System. Carbon. 2003;41(13):2607–2611. doi:10.1016/S0008-6223(03)00336-1

13. Zhou Z, Ci L, Chen X, et al. Controllable Growth of Double Wall Carbon Nanotubes in a Floating Catalytic System. Carbon. 2003;41(2):337–342. doi:10.1016/S0008-6223(02)00295-6

14. Nasibulin AG, Moisala A, Jiang H, Kauppinen EI. Carbon Nanotube Synthesis from Alcohols by a Novel Aerosol Method. J Nanopart Res. 2006;8(3):465–475. doi:10.1007/s11051-005-9027-8

15. Pinault M, Pichot V, Khodja H, Launois P, Reynaud C, Mayne-L’Hermite M. Evidence of Sequential Lift in Growth of Aligned Multiwalled Carbon Nanotube Multilayers. Nano Lett. 2005;5(12):2394–2398. doi:10.1021/nl051472k

16. Zhang Q, Wang D-G, Huang J-Q, et al. Dry Spinning Yarns from Vertically Aligned Carbon Nanotube Arrays Produced by an Improved Floating Catalyst Chemical Vapor Deposition Method. Carbon. 2010;48(10):2855–2861. doi:10.1016/j.carbon.2010.04.017

17. Barnard JS, Paukner C, Koziol KK. The Role of Carbon Precursor on Carbon Nanotube Chirality in Floating Catalyst Chemical Vapour Deposition. Nanoscale. 2016;8(39):17262–17270. doi:10.1039/C6NR03895F

18. Lekawa-Raus A, Patmore J, Kurzepa L, Bulmer J, Koziol K. Electrical Properties of Carbon Nanotube Based Fibers and Their Future Use in Electrical Wiring. Adv Funct Mater. 2014;24(24):3661–3682. doi:10.1002/adfm.201303716

19. Taborowska P, Giżewski T, Patmore J, Janczak D, Jakubowska M, Lekawa-Raus A. Spun Carbon Nanotube Fibres and Films as an Alternative to Printed Electronic Components. Materials. 2020;13(2):431. doi:10.3390/ma13020431

20. Sundaram RM, Koziol KKK, Windle AH. Continuous Direct Spinning of Fibers of Single-Walled Carbon Nanotubes with Metallic Chirality. Adv Mater Weinheim. 2011;23(43):5064–5068. doi:10.1002/adma.201102754

21. Paukner C, Koziol KKK. Ultra-Pure Single Wall Carbon Nanotube Fibres Continuously Spun without Promoter. Sci Rep. 2014;4(1):3903. doi:10.1038/srep03903

22. Dong L, Park JG, Leonhardt BE, Zhang S, Liang R. Continuous Synthesis of Double-Walled Carbon Nanotubes with Water-Assisted Floating Catalyst Chemical Vapor Deposition. Nanomaterials. 2020;10(2):365. doi:10.3390/nano10020365

23. Koziol K, Vilatela J, Moisala A, et al. High-Performance Carbon Nanotube Fiber. Science. 2007;318(5858):1892–1895. doi:10.1126/science.1147635

24. Hou P-X, Zhang F, Zhang L, Liu C, Cheng H-M. Synthesis of Carbon Nanotubes by Floating Catalyst Chemical Vapor Deposition and Their Applications. Adv Funct Mater. 2022;32(11):2108541. doi:10.1002/adfm.202108541

25. Boncel S, Pattinson SW, Geiser V, Shaffer MSP, Koziol KKK. En Route to Controlled Catalytic CVD Synthesis of Densely Packed and Vertically Aligned Nitrogen-Doped Carbon Nanotube Arrays. Beilstein J Nanotechnol. 2014;5(1):219–233. doi:10.3762/bjnano.5.24

26. Sen R, Govindaraj A, Rao CNR. Carbon Nanotubes by the Metallocene Route. Chem Phys Lett. 1997;267(3):276–280. doi:10.1016/S0009-2614(97)00080-8

27. Singh C, Quested T, Boothroyd CB, et al. Synthesis and Characterization of Carbon Nanofibers Produced by the Floating Catalyst Method. J Phys Chem B. 2002;106(42):10915–10922. doi:10.1021/jp026159a

28. Chauhan D, Pujari A, Zhang G, Dasgupta K, Shanov VN, Schulz MJ. Effect of a Metallocene Catalyst Mixture on CNT Yield Using the FC-CVD Process. Catalysts. 2022;12(3):287. doi:10.3390/catal12030287

29. Karaeva AR, Urvanov SA, Kazennov NV, Mitberg EB, Mordkovich VZS. Structure and Electrical Resistivity of Carbon Nanotubes Synthesized over Group VIII Metallocenes. Nanomaterials. 2020;10(11):2279. doi:10.3390/nano10112279

30. Li Y-L, Kinloch IA, Windle AH. Direct Spinning of Carbon Nanotube Fibers from Chemical Vapor Deposition Synthesis. Science. 2004;304(5668):276–278. doi:10.1126/science.1094982

31. Motta M, Moisala A, Kinloch IA, Windle AH. High Performance Fibres from ‘Dog Bone’ Carbon Nanotubes. Adv Mater. 2007;19(21):3721–3726. doi:10.1002/adma.200700516

32. Kim NS, Lee YT, Park J, et al. Dependence of the Vertically Aligned Growth of Carbon Nanotubes on the Catalysts. J Phys Chem B. 2002;106(36):9286–9290. doi:10.1021/jp021018u

33. Lim YD, Avramchuck AV, Grapov D, et al. Enhanced Carbon Nanotubes Growth Using Nickel/Ferrocene-Hybridized Catalyst. ACS Omega. 2017;2(9):6063–6071. doi:10.1021/acsomega.7b00858

34. Bronikowski MJ. CVD Growth of Carbon Nanotube Bundle Arrays. Carbon. 2006;44(13):2822–2832. doi:10.1016/j.carbon.2006.03.022

35. Kinloch IA, Shaffer MSP, Lam YM, Windle AH. High-Throughput Screening for Carbon Nanotube Production. Carbon. 2004;42(1):101–110. doi:10.1016/j.carbon.2003.10.004

36. Gommes C, Blacher S, Bossuot C, Marchot P, Nagy J, Pirard J-P. Influence of the Operating Conditions on the Production Rate of Multi-Walled Carbon Nanotubes in a CVD Reactor. Carbon. 2004;42(8):1473–1482. doi:10.1016/j.carbon.2004.01.063

37. Kim D-H, Jang H-S, Kim C-D, et al. Dynamic Growth Rate Behavior of a Carbon Nanotube Forest Characterized by in Situ Optical Growth Monitoring. Nano Lett. 2003;3(6):863–865. doi:10.1021/nl034212g

38. Yun Y, Shanov V, Tu Y, Subramaniam S, Schulz MJ. Growth Mechanism of Long Aligned Multiwall Carbon Nanotube Arrays by Water-Assisted Chemical Vapor Deposition. J Phys Chem B. 2006;110(47):23920–23925. doi:10.1021/jp057171g

39. Dresselhaus MS, Dresselhaus G, Saito R, Jorio A. Raman Spectroscopy of Carbon Nanotubes. Phys Rep. 2005;409(2):47–99. doi:10.1016/j.physrep.2004.10.006

40. Saito R, Hofmann M, Dresselhaus G, Jorio A, Dresselhaus MS. Raman Spectroscopy of Graphene and Carbon Nanotubes. Adv Phys. 2011;60(3):413–550. doi:10.1080/00018732.2011.582251

41. Dresselhaus MS, Dresselhaus G, Jorio A, Souza Filho AG, Saito R. Raman Spectroscopy on Isolated Single Wall Carbon Nanotubes. Carbon. 2002;40(12):2043–2061. doi:10.1016/S0008-6223(02)00066-0

42. Thostenson ET, Ren Z, Chou T-W. Advances in the Science and Technology of Carbon Nanotubes and Their Composites: a Review. Compos Sci Technol. 2001;61(13):1899–1912. doi:10.1016/S0266-3538(01)00094-X

43. Kaniyoor A, Ramaprabhu S. A Raman Spectroscopic Investigation of Graphite Oxide Derived Graphene. AIP Adv. 2012;2(3):032183. doi:10.1063/1.4756995

44. Subrahmanyam KS, Vivekchand SRC, Govindaraj A, Rao CNR. A Study of Graphenes Prepared by Different Methods: characterization, Properties and Solubilization. J Mater Chem. 2008;18(13):1517–1523. doi:10.1039/B716536F

45. Magedov IV, Frolova LV, Ovezmyradov M, Bethke D, Shaner EA, Kalugin NG. Benzyne-Functionalized Graphene and Graphite Characterized by Raman Spectroscopy and Energy Dispersive X-Ray Analysis. Carbon. 2013;54:192–200. doi:10.1016/j.carbon.2012.11.025

46. Kudin KN, Ozbas B, Schniepp HC, Prud’homme RK, Aksay IA, Car R. Raman Spectra of Graphite Oxide and Functionalized Graphene Sheets. Nano Lett. 2008;8(1):36–41. doi:10.1021/nl071822y

47. Al-Mashat L, Shin K, Kalantar-zadeh K, et al. Graphene/Polyaniline Nanocomposite for Hydrogen Sensing. J Phys Chem C. 2010;114(39):16168–16173. doi:10.1021/jp103134u

48. Abdelsayed V, Moussa S, Hassan HM, Aluri HS, Collinson MM, El-Shall MS. Photothermal Deoxygenation of Graphite Oxide with Laser Excitation in Solution and Graphene-Aided Increase in Water Temperature. J Phys Chem Lett. 2010;1(19):2804–2809. doi:10.1021/jz1011143

49. Sun S, Wu P. Competitive Surface-Enhanced Raman Scattering Effects in Noble Metal Nanoparticle-Decorated Graphene Sheets. Phys Chem Chem Phys. 2011;13(47):21116–21120. doi:10.1039/C1CP22727K

50. Ramansystems. Welcome to Ramansystems – Raman Frequencies. Available from: http://ramansystems.com/english/raman%20frequencies.htm. Accessed Dec 1, 2021.).

51. Janczak D, Wójkowska K, Raczyński T, et al. Development of Highly Stretchable Ag-MWCNT Composite for Screen-Printed Textile Electronics with Improved Mechanical and Electrical Properties. NSA. 2024;17:289–302. doi:10.2147/NSA.S493579

52. Janczak D, Słoma M, Wróblewski G, Młożniak A, Jakubowska M. Screen-Printed Resistive Pressure Sensors Containing Graphene Nanoplatelets and Carbon Nanotubes. Sensors. 2014;14(9):17304–17312. doi:10.3390/s140917304

53. Yáñez-Sedeño P, Pingarrón JM, Riu J, Rius FX. Electrochemical Sensing Based on Carbon Nanotubes. TrAC Trends in Analytical Chemistry. 2010;29(9):939–953. doi:10.1016/j.trac.2010.06.006

54. Saleh Ahammad AJ, Lee -J-J, Rahman Md A. Electrochemical Sensors Based on Carbon Nanotubes. Sensors. 2009;9(4):2289–2319. doi:10.3390/s90402289

55. Olabi AG, Abdelkareem MA, Wilberforce T, Sayed ET. Application of Graphene in Energy Storage Device – a Review. Renewable Sustainable Energy Rev. 2021;135:110026. doi:10.1016/j.rser.2020.110026

56. Chen T, Dai L. Carbon Nanomaterials for High-Performance Supercapacitors. Mater Today. 2013;16(7):272–280. doi:10.1016/j.mattod.2013.07.002

57. Zhu S, Sheng J, Chen Y, Ni J, Li Y. Carbon Nanotubes for Flexible Batteries: recent Progress and Future Perspective. Natl Sci Rev. 2020;8(5):nwaa261. doi:10.1093/nsr/nwaa261

58. Pattinson SW, Prehn K, Kinloch IA, et al. The Life and Death of Carbon Nanotubes. RSC Adv. 2012;2(7):2909–2913. doi:10.1039/C2RA00660J

Google Search is on the verge of its biggest overhaul yet, with “AI Mode” poised to become the new default.

AI Mode builds on Google’s existing “AI Overviews.” Like ChatGPT or Perplexity, it lets users keep chatting with the AI after an initial search, turning Google from a gateway to the web into a closed platform where generative AI is the main focus.

Google product manager Logan Kilpatrick has been hinting on X that AI Mode could soon become the default search experience. For example, google.com/ai now redirects directly to AI Mode, and when users asked for AI Mode to become the default, Kilpatrick replied with “soon : )”

AI Mode has launched in over 180 countries outside the EU, but it hasn’t replaced standard search yet. In its latest update, Google added new agent features to AI Mode, allowing users to book local services or buy event tickets directly through the chat interface.

Google’s lawyers say “the open web is already in rapid decline”

At the same time, Google’s legal team is arguing in a lawsuit over its ad tech division that, “The fact is that today, the open web is already in rapid decline” (via Jason Kint). They claim that breaking up Google’s ad business would make things worse, especially as ad dollars are already shifting away from traditional display ads on open websites and toward formats like connected TV, retail media, or what Google calls “enormously popular publishers, such as AI chatbots, who can monetize their display content.”

According to Google, forcing a split would only accelerate this shift, pulling resources out of the open web and hurting publishers who depend on that revenue. The company is pushing back against demands in the lawsuit to sell its AdX ad exchange, make its auction logic open source, and give up 50 percent of net revenue. Other proposals would require new rules of conduct and open APIs to rein in Google’s market power and increase competition.

The irony is hard to ignore: Google uses the decline of the open web as an argument to position itself as a defender of web publishers, even as it rolls out AI Mode and channels search traffic to its own platform, fueling the very trend it warns against.

In the end, Google is using a problem it played a major role in creating as a reason to fight efforts to break up its business. It’s a kind of circular logic that shows up in other contradictory statements from Google executives about how AI search affects the open web.

As most of us predicted from the moment we first saw Ninja Gaiden 4, it turns out that PlatinumGames is basically the main developer, with Team Ninja acting as advisors.

For the past decade and a bit, Ninja Gaiden fans have been begging Team Ninja to finally return to the series and give us a game that’s so good it makes us forget about the pre-Razor’s Edge version of Ninja Gaiden 3. Earlier this year, that wish was seemingly granted when Ninja Gaiden 4 was revealed out of nowhere.

Surprisingly, though, Ninja Gaiden 4 doesn’t just come from Team Ninja, as it’s actually a collaboration with PlatinumGames (as well as being funded by Xbox). Since that was confirmed, fans have wondered how the project was split up, with many believing that Ryu’s sections were handled by Team Ninja and Yakumo was all PlatinumGames.

Ninja Gaiden 4’s Main Development Was “Entirely” Done By PlatinumGames

As If That Wasn’t Already Clear From Yakumo

As pointed out by Redditor luneth22 on the Ninja Gaiden subreddit, that isn’t quite the case. Eurogamer recently had the chance to chat with Ninja Gaiden 4’s producers, including PlatinumGames’ Yuji Nakao and Team Ninja’s Masakazu Hirayama, who revealed exactly what each team provided for the game.

When I previewed Ninja Gaiden 4, I noticed that the intro credits specifically say that it’s “based on a concept by Team Ninja” but developed by Platinum.

When asked by Eurogamer how much of Ninja Gaiden 4 is Team Ninja and how much is Platinum, Hirayama revealed that the “main development” of the game takes place “entirely” at PlatinumGames. Team Ninja is sent daily builds of Ninja Gaiden 4 and then advises Platinum on what should be adjusted.

Of course, Team Ninja’s contributions are still very important, but it basically confirms once and for all that Ninja Gaiden 4 is much more of a Platinum game than you might think. Having played Ninja Gaiden 4 for a good few hours, though, I can confirm that it feels exactly how you’d want a Ninja Gaiden game to feel.

Hirayama also revealed that the reason Platinum was chosen as the developer for Ninja Gaiden 4 is because of its work on games like Nier: Automata and Bayonetta. Team Ninja’s producer noted that the collaboration is basically the only reason Ninja Gaiden 4 exists, as he notes that Team Ninja couldn’t have “achieved many things on our own”.

Oppo unveiled the Oppo K13 Turbo and Oppo K13 Turbo Pro in China in July, which were later introduced in India. Both smartphones have similar designs and specs, with the most significant difference being the chipset. We’ve received the Oppo K13 Turbo Pro, so let’s take a closer look at this mid-range smartphone.

Oppo K13 Turbo 5G

The Oppo K13 Turbo Pro comes in an attractive purple-colored retail box, which includes a SIM ejector tool, some documents, a protective case, a charging cable, and an 80W power adapter.

The Oppo K13 Turbo Pro has a 6.8″ 120Hz 2,800×1,280-pixel resolution LTPS AMOLED display on the front, which is protected by AGC DT-Star D+ glass. The panel has 10-bit color depth, a 240Hz touch sampling rate, and a peak brightness of 1,600 nits.

The K13 Turbo Pro’s screen supports 30Hz, 60Hz, 90Hz, and 120Hz refresh rates, and you’ll find three refresh rate modes in the phone’s display settings: Auto-select, Standard, and High. Auto-select automatically chooses the refresh rate to strike a balance between battery efficiency and a smooth experience. Standard mode refreshes the screen at 60Hz, while High mode goes up to 120Hz. There’s also an “App-specific refresh rate” option that lets you choose the desired refresh rate on a per-app (and game) basis.

Additionally, the Oppo K13 Turbo Pro’s display has support for wet touch and Glove Mode. The former lets you use the phone when there’s water on your fingers or the phone’s screen, whereas the latter allows you to operate the phone’s screen while wearing gloves that are not thicker than 5mm and are made of materials such as cotton, wool, and sheepskin. You can head over to the Settings > Accessibility & convenience menu to enable Glove Mode.

Talking about multimedia, the Oppo K13 Turbo’s screen supports Dolby Vision, HDR10, HDR10+, and HLG codecs, and it also has Widevine L1 certification, which allows 1080p video streaming in supported apps. It’s also worth noting that while the Oppo K13 Turbo Pro comes with an Always-On Display feature, it’s not always on since the screen only lights up and shows information when the phone moves. We’ve noticed this on some other Oppo smartphones as well, and it’s about time Oppo fixed it.

Moving on, the Oppo K13 Turbo Pro’s display has a punch hole in the center for the 16MP selfie camera. It uses a Sony IMX480 sensor and has an f/2.45 aperture and an 85° FOV. The front camera can record videos in up to 1080p resolution at 30 fps.

Underneath the screen is an optical fingerprint reader. It was easy to set up, and we found it to be fast and accurate during our time with the phone. However, we’d have liked it to be placed higher for a more convenient phone-unlocking experience.

Flip the Oppo K13 Turbo Pro around, and you see a squircle-shaped camera island aligned vertically in the top-left corner of the phone. At first glance, it might look like the K13 Turbo Pro has three cameras on the rear, accompanied by an LED flash, but that’s not the case since the smartphone features two rear cameras.

The first circle inside the camera island is for the primary camera, while the second circle is for the built-in cooling fan, which is surrounded by dual Mist Shadow breathing LEDs (more on that later). The second camera is located on the right side of the island, below which is the flash.

The Oppo K13 Turbo Pro’s primary camera, having OIS and the capability to record videos in up to 4K resolution at 60 fps, uses an OmniVision OV50D40 sensor and has an f/1.8 aperture. The second camera is a monochrome unit with an f/2.4 aperture and uses OmniVision’s OV02B1B sensor.

Oppo K13 Turbo Pro with cooling fan and dual Mist Shadow breathing LEDs on its back

The Oppo K13 Turbo Pro also comes with some AI-powered camera features, including AI Clarity Enhancer, AI Unblur, AI Reflection Remover, and AI Eraser 2.0.

The K13 Turbo Pro comes in Silver Knight, Purple Phantom, and Midnight Maverick colors. The Midnight Maverick model features a rear cover with a clean look devoid of any patterns, while the other two versions have identical designs on their back covers.

The subject of this hands-on is the Silver Knight model, which, Oppo says, is “inspired by the rugged metal of high-speed motorcycles and their gleaming silver exhausts.” Its back cover has a matte finish and does a good job at resisting smudges. It also has “Active Cooling” and “Master The Wind” written on it.

The Oppo K13 Turbo Pro’s Silver Knight version looks cool, and while it doesn’t give you a premium in-hand feel like expensive flagships, it doesn’t feel cheap either. We like the overall build. However, with a 6.8″ screen, the smartphone could be unwieldy for some for one-handed usage. Besides, the author of this article would’ve also preferred the phone’s back panel with some curvature for a more comfortable in-hand experience. Some curves around the edges would’ve been nice.

That said, the Oppo K13 Turbo Pro has a USB-C port at the bottom, flanked by a speaker grille, a microphone, and a SIM card slot. At the top is the IR blaster joined by another mic. On the right side is the power button and volume rocker, above which is a duct for releasing the hot air.

Ports and controls

This brings us to the biggest highlight of the Oppo K13 Turbo Pro: the built-in active cooling fan, which, like the smartphone, is IPX6, IPX8, and IPX9-rated. The in-built cooling fan is meant for two things: 1) to improve the K13 Turbo Pro’s fall resistance with its stacking design; 2) to help keep the phone cool.

The Oppo K13 Turbo Pro’s built-in fan has 0.1mm blades, capable of reaching a rotation speed of up to 18,000 RPM, which increases air volume by 20%, while the Arc-shaped vortex tongue design boosts air intake by about 10%. Furthermore, Oppo claims that the smartphone’s 13 ultra-thin 0.1mm fins with uneven spacing improve heat dissipation efficiency by 3 times compared to traditional solutions, while the L-shaped duct with 3mm design minimizes airflow resistance and boosts cooling efficiency for peak performance. This active cooling system is advertised to help the K13 Turbo Pro run 2-4℃ cooler under high-load conditions.

Oppo K13 Turbo has a cooling fan on the rear and a duct on the right frame

However, the Oppo K13 Turbo Pro doesn’t rely on the built-in fan alone to stay cool, as it also comes with a 7,000mm₂ vapour chamber for heat dissipation. With all that cooling tech put to work, Oppo said the K13 Turbo Pro “consistently achieved the lowest overall temperature among competitors during a 3-hour gaming session at 25°C and 120 FPS” in its internal lab tests.

Well, we ran some benchmark tests on the Oppo K13 Turbo Pro and did some gaming with and without the built-in cooling fan turned on to see if it actually makes any difference in real-world usage, but before we delve into that, let’s quickly check what customization options you get for the cooling fan and the dual Mist Shadow breathing LEDs on the rear.

The Oppo K13 Turbo Pro’s cooling fan automatically turns on when you open a game or AnTuTu, even if it’s disabled, and when the smartphone is running hot, the system alerts you to turn it on. You can either do that from the notification itself or by navigating to the K13 Turbo Pro’s Settings > Cooling fan menu.

Under this menu, you’ll find that the cooling fan supports two cooling modes – Smart shifting and Full speed. You can also choose to have the fan automatically turned on for games, outdoor mode, and when the phone is fast charging.

The cooling fan supports four startup sound effects – Charge, Storm, Battle, and Howl. You can also choose to have no sound at all, or pick a custom ringtone if that’s more your style.

Additionally, you can choose to enable or disable the RGB lights on the back. By default, they are enabled, and they support eight color effects. The lights turn on when the fan is spinning.

It’s also worth mentioning that the cooling fan only spins when the phone’s screen is on. Once the screen is off, the fan stops spinning. We’d like Oppo to change that and allow users to keep the fan spinning even when the screen is off so that the phone can cool down faster. This would make the built-in fan more useful.

With those cooling fan customizations out of the way, let’s talk about the phone and the fan’s performance now.

The Oppo K13 Turbo Pro is powered by the Snapdragon 8s Gen 4 SoC and comes with up to 12GB of LPDDR5X RAM and 256GB of UFS 4.0 storage. Our unit has 12GB RAM and 256GB storage onboard. It boots ColorOS 15 based on Android 15 and comes with the promise of two years of OS upgrades and three years of security updates.

The K13 Turbo Pro comes pre-loaded with more than 15 third-party apps and games. Fortunately, all of these can be uninstalled, but the number is on the higher side, and it’s time Oppo sold smartphones with fewer pre-loaded apps and games.

We ran AnTuTu 10, Geekbench 6, and 3DMark on the Oppo K13 Turbo Pro, and the results were similar to what we saw on a couple of other Snapdragon 8s Gen 4-powered smartphones that we tested, except for 3DMark’s Solar Bay scores, which were unusually low. This could be due to a software glitch.

These benchmarks were run with and without the built-in cooling fan turned on, and there were no significant differences in the test numbers or the phone temperature reported by the system. In fact, the scores were always lower when the fan was turned on. You can check the charts below for more details.

The story’s no different with the 1-hour CPU Throttling Test since the phone’s CPU always throttled to 60-64% regardless of the cooling fan being on or off.

With cooling fan turned off • With cooling fan turned on

Where it shined, though, was in 3DMark’s 20-minute Wild Life Extreme Stress Test. With the cooling fan turned off, the benchmark reported a stability of around 50%. But with the fan turned on, it was always above 75%, with the best score being 83.3%. For what it’s worth, there was also a difference of 1°C in temperature with the fan spinning, even though the phone itself was very hot to the touch.

With cooling fan turned off • With cooling fan turned on

Talking about real-world usage, we ran some popular gaming titles continuously for an hour at the highest graphics setting supported, with and without the cooling fan turned on. The performance remained smooth in both scenarios, with the frame rate being steady throughout the gameplay. The area around the camera module did get warm with the fan turned off, but not to the point where it would be uncomfortable to hold the phone. With the fan spinning, the phone always remained about 2°C cooler after an hour-long gaming session.

Note that we set the cooling mode of the fan to Full speed instead of Smart shifting when playing games and running benchmarks for the sake of consistency.

Keeping the Oppo K13 Turbo Pro up and running is a 7,000 mAh battery with 80W charging support. It also supports bypass charging. Oppo claims the battery can go from 1% to 68% in 30 minutes and 100% in 54 minutes with the bundled 80W power adapter.

In our testing, the battery charged from 1% to 13% in 5 minutes, 33% in 15 minutes, 50% in 25 minutes, 58% in 30 minutes, and 100% in 58 minutes. Your mileage will likely vary depending on your usage and ambient temperature when charging the phone.

We couldn’t run our battery life test on the Oppo K13 Turbo Pro since it cannot be run outside our HQ, but with a capacity of 7,000 mAh, it should get you through the day with ease.

The Oppo K13 Turbo Pro is available in India in Midnight Maverick, Purple Phantom, and Silver Knight colorways with two memory options – 8GB/256GB and 12GB/256GB, priced at INR37,999 ($430/€370) and INR39,999 ($455/€390), respectively.

In China, its base model has 12GB RAM and 256GB storage, and it’s priced at CNY1,999 ($280/€240/INR24,665).