
pmid: 21538588
Electronic and optoelectronic semiconductor components are the building blocks of modern instrumentation and equipment for sensing, computation, display, and communication. Systems incorporating these components are typically made on mechanically rigid printed circuit boards (PCBs). These systems can also be built on polymer-based fl exible PCBs, [ 1 ] which offer a bending radius of several centimeters about a single axis but are subject to fracturing from excessive bending or fatigue strain. Systems that are highly bendable (millimeter scale), stretchable, conformable to any surface topology, and mechanically insensitive to fatigue strain would greatly expand the application space of electronics. For example, in medicine there is a need for electronics to conform to and deform with the human body [ 2 ] to perform accurate diagnosis and deliver therapy. Other application spaces include renewable energy, [ 3,4 ] robotics, [ 5 ] military, [ 6 ] and lighting. [ 7 ] These applications have motivated research in fl exible and/or stretchable organic electronics [ 6 , 8–9 ] and inorganic electronics assembled on stretchable substrates. [ 2,4 , 7 , 10 ] One approach to building stretchable inorganic electronics is to connect thin electronic components together with stretchable spring-like metal interconnects and embed the entire interconnected structure into a stretchable (rubber) substrate. [ 2–3 , 7 ] Whereas prior work based on this approach used custom microfabrication of the electronic components and interconnects, here we present a process that uses commercially available electronic components and fl ip-chip bonding processes. Therefore, this fabrication process is a platform that can be used without modifi cation to create stretchable electronic systems incorporating any set of electronic components. As a demonstration, we fabricated stretchable light-emitting diode (LED) arrays containing up to 50 LEDs and show that the arrays can survive repeated stretching of 90 000 cycles and also tightly conform to a human thumb tip. The general concept behind the fabrication process is to separately manufacture the electronic components and the stretchable interconnects, then combine them using fl ip-chip bonding technology. We term this process “CINE” (combination of interconnects and electronics). Specifi cally, the process involves three steps: 1) the fabrication of metal contact pads and stretchable interconnects using standard microfabrication techniques; 2) transfer printing the contact pads and stretchable interconnects to a stretchable substrate using dissolvable adhesives as the intermediate transfer material; and 3) fl ip-chip bonding the electronic components onto the metal contact pads using anisotropic conductive fi lm (ACF). A detailed description of the fabrication process is presented in the Experimental Section. Figure 1 shows a stretchable LED array fabricated with this process. The stretchable LED array consists of ten pairs of gold contact pads, connected by serpentine-shaped metal interconnects. The interconnects are fully encapsulated in polyimide, whereas the contact pads have openings to allow electrical contact. Five blue and fi ve red LEDs were fl ip-chip bonded to the pairs of contact pads (in opposing polarities so the array can be powered with either a positive or negative voltage bias). The array was made on a silicone substrate with an elastic modulus of 10 kPa (we used the material EcoFlex made by Smooth-On Inc.). When the array is stretched, the serpentine-shaped interconnects deform to accommodate most of the strain, minimizing the strain seen by the LEDs and allowing the LEDs to maintain their optoelectronic properties (Figure 1 c,e). When the substrate is stretched, the interconnects accommodate strain via out-of-plane buckling as well as lateral deformation; this out-of-plane deformation is possible because the EcoFlex substrate is extremely compliant. We measured the resistance of individual interconnects while being stretched and found no signifi cant change in electrical resistance. To test the mechanical robustness of the arrays, we repeatedly stretched them in the length-wise direction using a mechanical actuator (additional details are provided in the Supporting Information). In the initial state, the array was without any strain, and we measured the distance between two adjacent LEDs as L 0 . When the array was fully stretched, we remeasured the distance between the two adjacent LEDs as L 1 . We defi ne the strain as ( L 1 – L 0 )/ L 0 . With a peak stretching strain of 67%, the arrays survived up to 90 000 stretching cycles (at an oscillating frequency of 1 Hz). With a peak stretching strain of 200%, the arrays survived up to 5000 cycles. We determined the failure mechanism to consistently be fracture–breakage of the serpentine interconnects near the contact pads used for powering the array. These contact pads were too large, at about 1 cm 2 in size, and created regions of high localized strain around their edges. The interconnects connecting adjacent LEDs never failed and neither did the fl ip-chip bonds made between the LEDs and the contact pads. Therefore, we expect the mechanical robustness of the arrays to dramatically increase simply by redesigning the end-most contact pads to be smaller by a factor of about four. To examine the electrical robustness, we measured the current-voltage ( I – V ) relation of an LED array prior to being stretched, after being stretched 1000 cycles, and after being stretched 10 000 cycles. We found no signifi cant variation in the I – V characteristics; the results are presented in Figure 2 a. We also measured the current fl owing through an array of 15 LEDs (arranged as three parallel sets of fi ve LEDs in-series), biased at 20 V. The current fl ow was a constant 93 mA as the
Silicon, Semiconductors, Inorganic Chemicals, Gold, Electronics
Silicon, Semiconductors, Inorganic Chemicals, Gold, Electronics
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 145 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Top 1% | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Top 1% | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Top 1% |
