研究背景
柔性/可穿戴电子器件因其在经历巨大变形时仍能可靠地运作，以及其广泛的应用，包括电磁屏蔽、智能显示器、电子皮肤、软体机器人、人机界面和植入物等，现在正引起人们的极大关注。例如，具有良好透气性的可穿戴电子器件可以舒适地穿戴在人体皮肤上，以捕捉电生理信号，如脉搏、呼吸、和汗水，这可以识别潜在的健康危害。尽管最近在增材制造方面取得了进展，但传统的方法，如溅射、蒸发沉积和无电解镀工艺仍有一些缺点，包括多阶段的程序、高成本的设备和危险废物的产生。印刷电子技术的出现和改进促进了柔性/可拉伸电子技术的制造。据报道，2020 年印刷电子的市场规模已经达到 78 亿美元，预计到 2026 年将上升到 207 亿美元。例如，印刷电子领域的柔性混合电子(FHE)正在迅速发展和广泛使用。FHE利用了硅集成电路和柔性印刷技术的不同优势。因此，它在许多新兴领域有更广泛的应用，如可穿戴医疗设备和工业及农业传感。
传统的印刷和复制行业现在正逐渐向数字化、网络化和一体化转变。最近，在印刷方法中，喷墨印刷作为一种新兴的技术引起了越来越多的关注，它具有卓越的数字化能力和高沉积分辨率，在过去的半个世纪中得到了飞速的发展。喷墨印刷是一种完全数字化的印刷过程，没有传统印刷过程中的繁琐程序，因此，它具有高的时间效率和低的成本。在打印过程中，打印喷嘴由计算机控制并进行平移运动，而且由于对液滴沉积的精心控制和基于纳米材料(如金属、二维(2D)材料、钙钛矿和生物材料的功能性油墨的引入。喷墨打印能够在各种基材上实现厚度从几纳米到几十微米的结构，如纸张、聚合物、纺织品、陶瓷和皮肤，具有很高的适应性，并在能源转化、传感、显示、生物医学等方面显示出巨大的潜力 (图1)。尽管喷墨打印在打印分辨率和可扩展性方面并不突出，但它在这些参数之间保持了综合平衡。用于喷墨打印的墨水的粘度约为 1-20 厘泊，与水的粘度相似，以保持良好的喷墨性能。与现有的商业印刷方法相比，这些优势赋予了喷墨印刷更高的效率、更好的印刷质量、更短的生产周期、更丰富的信息来源和更容易的信息传输。喷墨打印技术为制造可穿戴电子产品提供了若干优势，如轻质、环保和具有成本效益的生产过程，这符合“真正的”可穿戴应用的需要。
研究成果
柔性/可穿戴电子设备的快速发展需要新的制造策略。在最先进的技术中，喷墨打印引起了相当大的兴趣，因为它可以大规模地制造具有良好可靠性、高时间效率和低制造成本的柔性电子器件。在这篇综述中，基于工作原理，总结了喷墨打印技术在柔性/可穿戴电子领域的最新进展，包括柔性超级电容器、晶体管、传感器、热电发生器、可穿戴织物，以及用于射频识别。此外，还讨论了该领域目前的一些挑战和未来的机遇。希望这篇评论文章能给柔性电子学领域的研究人员提供积极的建议。青岛大学Yu Liu& Hongze Zhu为文章的第一作者，青岛大学孙彬副教授为通讯作者。相关报道以“Recent advances in inkjet-printing technologies for flexible/wearable electronics”为题发表在Nanoscale期刊上。
综述要点
1. 基于喷墨打印的工作原理，总结了喷墨打印技术的最新进展，
2. 讨论了喷墨打印在柔性电子领域的应用，包括超级电容、晶体管、传感器、光电探测器、热电发生器、可穿戴织物、射频识别等。
3. 最后讨论了一些未来的挑战和实际应用的机会。
图文导读
Fig. 1 Schematic diagram of partial materials and applications of inkjet printing.
Fig. 2 Schematic diagram of (a) the continuous inkjet printing and (b) the drop-on-demand inkjet printing technique.
Fig. 3 (a) Schematic diagram of the thermal bubble printing principle. (b) Classification of piezoelectric inkjet printing.
Fig. 4 (a) Schematic diagram of the MSC production process. (b) CV Curves of inkjet printing MSCs under different bending radii. (c) Demonstration of ultra-thin, flexibility, and capacitance retention performance of inkjet printing MSCs under different bending radii. (d) Bending test of MSCs. (e) Relationship of capacitance retention and charge–discharge cycles.
Fig. 5 (a) Sketch of the intrinsically-stretchable array of transistors representing how each active material of the transistor was additively fabricated with the same inkjet printing method. (b) Sketch of the cross-section of a device, where each layer material is labeled. (c) Overlapped transfer curves of the inkjet printed SWCNT-FET subjected to different strain conditions along the two main directions: perpendicular (left) and parallel (right) to the channel length. (d) Picture of an array of transistors IJ printed over a large area and bent over a hand.
Fig. 6 (a) Three-dimensional view of the intrinsically stretchable transistor array as a core building block for skin electronics. (b and c) Mobilities and threshold voltages during a stretching cycle parallel (f) and perpendicular (g) to the channel direction. (d) A stretchable active matrix developed from our intrinsically stretchable transistor array. (e) Transfer curves from the inverter when stretched gradually from 0% to 100% strain.
Fig. 7 (a) IVL of p-PLEDs on glass substrates (black square), a-PLEDs on glass substrates (red circle), and a-PLEDs on PEN substrates (green triangle), (b) efficiencies of a-PLEDs on PEN substrates before (green triangle) and after 100 bending cycles (blue inverted triangle). (c) Transmission characteristics of a P-type SWCNT transistor. (d) Transmission characteristics of an N-type IO transistor. (e) Transmission characteristics of an anti-bipolar transistor with a channel consisting of an N-type IO and a P-type SWCNT.
Fig. 8 (a) All-graphene humidity sensors built on the PET substrate. (b) Device response before (left) and after (right) 2000 bends. (c) Inkjet-printed graphene-based flexible humidity sensor. (d) The measured capacitance of the humidity sensor during relative humidity from 30% RH to 90% RH at 25 °C. (e) The measured capacitance of the humidity sensor during the relative humidity from 30% RH to 90% RH at 40 °C.
Fig. 9 (a) Photograph of the printed temperature sensor (b) relative resistance changes of the printed temperature sensor concerning the bending cycles of up to 5000 cycles (5 mm). (c) Temperature-dependent relative resistance changes of the printed temperature sensor before bending (0 cycles) and after cyclic bending for 5000 cycles at a bending radius of 5 mm (5000 cycles). (d) The layout of temperature sensor design. (e) Digital image of the printed temperature sensor. (f) The resistance variations of the sensors folded under various curvature angles.
Fig. 10 (a) Schematic illustration of the fabrication of bimodal sensor and circuit diagram. (b) Gauge factor of the strain sensor under different strains. (c) Mechanical durability of the strain sensor under a strain of 0.76% and 1.07%, respectively. (d) The durability test of our pressure sensor under pressure of 202.8 kPa. (e) Real-time fast response (≈90 ms) of the PUA/Ag single-layer strain sensor under 1% strain. (f) Photo of a representative sample with multiple printed pressure sensors. (g) Relative change in resistance of the printed pressure sensor in response to acoustic vibrations under various sound pressure levels (SPL) played through a speaker with prerecorded voice.
Fig. 11 (a) Inkjet printing PEDOT:PSS line structures with control of drop spacing (DS). (b) Measured surface atomic force microscopy (AFM) images of the inkjet-printed PEDOT:PSS line structures with different drop spacing. (c) Measured ΔR/R0 as functions of the NH3 concentration for the different devices. (d) Photograph of the completed sensor matrix. (e) Schematic of the silver interdigital electrode with pitch (P), gap (G), and width (W) parameters. (f) Long-term response stability over one month to 3 ppm of PA for sensor unit.
Fig. 12 (a) Schematic illustration of inkjet printing of large-scale perovskite films. (b) UV-vis spectra of the perovskite films inkjet-printed with and without soluble PEO layer. (c) The current–voltage curve in different incident light intensities. (d) Current–time curves of the photodetector with different bending cycles. (e) The photocurrent along θ, in which each point is the average photocurrent from ϕ = 0° to ϕ = 360°. (f) The normalized photocurrent as a function of strain. (g) The performance of the photodetector as a function of the stretching cycle with a strain of 60%.
Fig. 13 (a) SEM image of the deposited copper on the modified paper substrate. (b) Optical image of RFID tag. (c) Normalized reading range of RFID tag antennas versus bending times. (d–f) Frequency response in terms of transmission coefficient S21 for the single-layer, mirror-stacked, and inline-stacked Miura-Ori assemblages for folding angles θ = 60°, 90°， 120° corresponding to panel extensions of 50%, 71%, and 87%, respectively, and considering a normal angle of incidence (AoI) with a linearly polarized electric field E that is oriented along the length of the dipole elements.
Fig. 14 (a) Schematic representation of the protein-mediated assembly of MXene sheets during inkjet printing. (b) Images of the LED circuit inkjet-printed on PET using ink P5 at (i) initial and (ii) bent states (images focused on the area of deformation are provided as insets). (c) The normalized voltage on LED as a function of the radius of curvature for the inkjet-printed electrodes (outline of the circuit design is provided as the inset). (d) Electromagnetic interference shielding efficiency (EMI SE) of large area (circle, d = 2 cm) electrodes inkjet-printed on PET using various ink solutions (MXene ink, ink P5) at 60% relative humidity conditions.
Fig. 15 (a) Photograph of the thermal heater (size: 35 × 35 mm) peeled off from the polyimide film. (b) Relationship between the number of bends of Au wiring printed on the PDLLA thin-film and the change in the resistance of Au wiring. (c) IR thermography of the thin-film heater attached to the hepatic lobe after wireless powering. (d) The cross-sectional view of histopathological dissection of the liver after 5 min of heat generation. (e) Photographs of an inkjet-printed device consisting of 20 silver and graphene legs bent (above) and as was (below). (f) Temperature-dependent post-annealing lattice (filled symbols) and electronic (open symbols) thermal conductivities. (g) Relative change in a single graphene leg thermoelectric performance as a function of bend cycles (error bars shown for power factor). (h) Relative change in a single graphene leg resistance as a function of bend radius under tensile (upper inset) and compressive (lower inset) bending as indicated by the colored arrows. (i) Output voltage and power of the TEG with 600 PN pairs versus current at various temperature differences. (j) Resistance and output voltage changes of the TEG versus the number of bending cycles at a bending radius of 18 mm. (k) Output voltage of the thermoelectric generator over time, by harvesting thermal heat from the human body. (l) Photographs showing a red light-emitting diode (LED) powered with the CNTY-based thermoelectric generator.
Fig. 16 (a and b) SEM images at different magnifications showing the silver nanoparticle pigment printed on CNFs/glycerol-coated woven cotton fabrics. (c) Manufacturing of inkjet-printed paths using silver nanoparticle ink on cotton-woven fabrics coated with CNFs/glycerol with connected LED and battery. (d) Transmission optical microscopy at 10×resolution of in situ heat-cured woven fabrics. Insert is the corresponding SEM images showing the silver-coated areas. (e) Change of the normalized resistance of the inkjet-printed conductive knit over the bending cycles. (f) Scanning electron microscopy (SEM) images of the L-Ti3C2-printed fabric surface with print passes 〈N〉 = 10. (g) L-Ti3C2Tx printed conductive circuit on cotton knit fabric with 〈N〉 = 1. (h) Areal capacitance as a function of scan rate, and (i) cycling stability over 10 000 cycles at 50 mV s−1.
Fig. 17 (a) Schematic of small-area perovskite solar cell (PSC) stack with deposition methods for each layer. (b) Current–voltage (J–V) characteristics of champion solar cell with inkjet-printed triple-cation perovskite absorber layer. (c) Transmittance (T) spectra and the corresponding Rs of Ag mesh@PDA/PET with different line widths, ITO/PET, and commercial Ag grids. (d) Current density–voltage (J–V) curves of Ag mesh@PDA/PET and ITO/PET-based flexible PSCs. (e) Corresponding normalized average PCE of flexible PSCs without encapsulation as a function of bending cycles (radius of 4 mm). (f) Corresponding normalized average PCE of flexible PSCs without encapsulation as a function of storage time.
总结与展望
喷墨打印技术在从光伏和显示技术到生物医学等多个领域显示出巨大潜力。特别是随着柔性/可穿戴电子设备的发展，由于工艺简单、加工时间短、材料消耗低、分辨率高、非接触式制造等特点，喷墨打印技术在实验室研究和工业制造方面都显示出大规模生产此类设备的巨大潜力，目前已被广泛研究。用于喷墨打印的油墨种类繁多，根据电功能的不同，大致可以分为几类。导电性油墨，如Ag， Au，PEDOT:PSS，rGO，MXene。由于导电性强，大多用于制作导电线和电极。像钛酸钡和钇氧化铝(YAIOx)这样的墨水可以作为介电墨水，在制造场效应晶体管和电容器中很常见。基于半导体的墨水，如金属氧化物、SWCNT、过氧化物和氧化钢，总是被用来制造晶体管、P-N结和传感器。尽管由于热发泡和压电喷墨打印机的性能限制，总是需要一定的粘度来确保墨水顺利喷出，但与其他打印方法相比，喷墨打印墨水的粘度(从1到 2P) 是最低的。在目前所有的印刷技术中，喷墨印刷中最小的点的尺寸也是最小的，这意味着与其他印刷技术相比，它具有最高的精确度。
喷墨打印可以在几乎所有的基材上执行。根据物理特性，基材也可分为几种类型。有松散和多孔的基质，如棉纱、聚酷织物、针织布、编织布和无纺布、编织棉布、碳纳米管纱(CNTY)；可拉伸的基质包括 PDMS、TPU；柔性基质如 PET、薄PEO 层和柔性 PEN以及最常用的基质如纸、和相纸。其他没有柔韧性和弹性的材料，如硬质玻璃、掺氟氧化锡涂层玻璃、和紫外线固化的 PUA基材也可以用作喷打印基材。有时需要对基材进行表面改性，以改善高孔隙率，减少墨水的渗透，增加墨水和基材之间的附着力。
然而，除了重大进展之外，喷墨打印作为一种新的新兴技术，仍然可以迎接巨大的挑战。首先，为制造多功能设备，应合成各种不同内容和功能的油墨，以满足纳米技术快速发展的需求。其次，应修正喷墨数字印刷的质量缺陷，其中包括色彩再现、纸张平整度、图形边缘锐利度、背面条纹、脏墨点和白线等， 尽管很高兴看到已经取得了一些成就，如在基材表面呈现功能性涂层。这种涂层可以精确调节一滴墨水的渗透和扩散之间的相对量。此外，由于不均匀的制造/固化、环境湿度/温度影响和不可控的沉积条件，特别是在低生产设置中存在着不足之处。通过机器学习模型来解释变量并输出高置信度的预测信号，可以使用合适的喷墨打印机以最小的生产努力来打印物体，同时还可以提取可靠的数据。这使得那些处于孤立/受限环境、贫困社区、资源匮乏的环境或业余爱好者可以在家里使用喷墨打印。此外，低分辨率、慢速打印、低产量以及可能的喷嘴堵塞等问题仍然存在。在光电探测器阵列中，每个光电探测器作为图像传感器的一个像素，最好在矩阵中嵌入更多的像素。尽管如此，喷墨打印相对较低的分辨率仍然阻碍了在这一领域的进一步应用。然而，电子产品的市场仍然很大，而且在不断增长，这意味着对喷墨打印工艺的理解和优化非常重要。除了柔性电子，喷墨打印还被广泛用于许多领域，如光伏、水净化和防伪技术，表明这项技术的前景十分广阔。
文献链接
Recent advances in inkjet-printing technologies for flexible/wearable electronics
DOI: 10.1039/d2nr05649f
转自：“i学术i科研”微信公众号
如有侵权，请联系本站删除！
