Chipmakers proceed to claw for each spare nanometer to proceed cutting down circuits, however a know-how involving issues which are a lot larger—a whole bunch or hundreds of nanometers throughout—might be simply as vital over the subsequent 5 years.
Called hybrid bonding, that know-how stacks two or extra chips atop each other in the identical package deal. That permits chipmakers to enhance the variety of transistors of their processors and recollections regardless of a normal slowdown within the shrinking of transistors, which as soon as drove Moore’s Law. At the
IEEE Electronic Components and Technology Conference (ECTC) this previous May in Denver, analysis teams from world wide unveiled a wide range of hard-fought enhancements to the know-how, with a number of displaying outcomes that would lead to a report density of connections between 3D stacked chips: some 7 million hyperlinks per sq. millimeter of silicon.
All these connections are wanted due to the brand new nature of progress in
semiconductors, Intel’s Yi Shi informed engineers at ECTC. Moore’s Law is now ruled by an idea known as system know-how co-optimization, or STCO, whereby a chip’s capabilities, comparable to cache reminiscence, enter/output, and logic, are fabricated individually utilizing the very best manufacturing know-how for every. Hybrid bonding and different superior packaging tech can then be used to assemble these subsystems in order that they work each bit in addition to a single piece of silicon. But that may occur solely when there’s a excessive density of connections that may shuttle bits between the separate items of silicon with little delay or vitality consumption.
Out of all of the advanced-packaging applied sciences, hybrid bonding supplies the best density of vertical connections. Consequently, it’s the quickest rising section of the advanced-packaging trade, says
Gabriella Pereira, know-how and market analyst at Yole Group. The total market is about to greater than triple to US $38 billion by 2029, in accordance to Yole, which tasks that hybrid bonding will make up about half the market by then, though as we speak it’s only a small portion.
In hybrid bonding, copper pads are constructed on the highest face of every chip. The copper is surrounded by insulation, often silicon oxide, and the pads themselves are barely recessed from the floor of the insulation. After the oxide is chemically modified, the 2 chips are then pressed collectively face-to-face, in order that the recessed pads on every align. This sandwich is then slowly heated, inflicting the copper to increase throughout the hole and fuse, connecting the 2 chips.
Hybrid bonding can both connect particular person chips of 1 measurement to a wafer stuffed with chips of a bigger measurement or bond two full wafers of chips of the identical measurement. Thanks partly to its use in digital camera chips, the latter course of is extra mature than the previous, Pereira says. For instance, engineers on the European microelectronics-research institute
Imec have created a number of the most dense wafer-on-wafer bonds ever, with a bond-to-bond distance (or pitch) of simply 400 nanometers. But Imec managed solely a 2-micrometer pitch for chip-on-wafer bonding.
The latter is a large enchancment over the superior 3D chips in manufacturing as we speak, which have connections about 9 μm aside. And it’s a good larger leap over the predecessor know-how: “microbumps” of solder, which have pitches within the tens of micrometers.
“With the equipment available, it’s easier to align wafer to wafer than chip to wafer. Most processes for microelectronics are made for [full] wafers,” says
Jean-Charles Souriau, scientific chief in integration and packaging on the French analysis group CEA Leti. But it’s chip-on-wafer (or die-to-wafer) that’s making a splash in high-end processors comparable to these from AMD, the place the approach is used to assemble compute cores and cache reminiscence in its superior CPUs and AI accelerators.
In pushing for tighter and tighter pitches for each eventualities, researchers are targeted on making surfaces flatter, getting sure wafers to stick collectively higher, and chopping the time and complexity of the entire course of. Getting it proper may revolutionize how chips are designed.
WoW, Those Are Some Tight Pitches
The current wafer-on-wafer (WoW) analysis that achieved the tightest pitches—from 360 nm to 500 nm—concerned loads of effort on one factor: flatness. To bond two wafers along with 100-nm-level accuracy, the entire wafer has to be practically completely flat. If it’s bowed or warped to the slightest diploma, complete sections gained’t join.
Flattening wafers is the job of a course of known as chemical mechanical planarization, or CMP. It’s important to chipmaking usually, particularly for producing the layers of interconnects above the
transistors.
“CMP is a key parameter we have to control for hybrid bonding,” says Souriau. The outcomes introduced at ECTC present CMP being taken to one other stage, not simply flattening throughout the wafer however lowering mere nanometers of roundness on the insulation between the copper pads to guarantee higher connections.
“It’s difficult to say what the limit will be. Things are moving very fast.” —Jean-Charles Souriau, CEA Leti
Other researchers targeted on making certain these flattened components stick collectively strongly sufficient. They did so by experimenting with totally different floor supplies comparable to silicon carbonitride as a substitute of silicon oxide and by utilizing totally different schemes to chemically activate the floor. Initially, when wafers or dies are pressed collectively, they’re held in place with comparatively weak hydrogen bonds, and the priority is whether or not every thing will keep in place throughout additional processing steps. After attachment, wafers and chips are then heated slowly, in a course of known as annealing, to type stronger chemical bonds. Just how robust these bonds are—and even how to determine that out—was the topic of a lot of the analysis introduced at ECTC.
Part of that closing bond power comes from the copper connections. The annealing step expands the copper throughout the hole to type a conductive bridge. Controlling the dimensions of that hole is essential, explains Samsung’s
Seung Ho Hahn. Too little growth, and the copper gained’t fuse. Too a lot, and the wafers shall be pushed aside. It’s a matter of nanometers, and Hahn reported analysis on a brand new chemical course of that he hopes to use to get it good by etching away the copper a single atomic layer at a time.
The high quality of the connection counts, too. The metals in chip interconnects are usually not a single crystal; as a substitute they’re made up of many grains, crystals oriented in several instructions. Even after the copper expands, the metallic’s grain boundaries usually don’t cross from one facet to one other. Such a crossing ought to cut back a connection’s electrical resistance and enhance its reliability. Researchers at Tohoku University in Japan reported a brand new metallurgical scheme that would lastly generate massive, single grains of copper that cross the boundary. “This is a drastic change,” says
Takafumi Fukushima, an affiliate professor at Tohoku. “We are now analyzing what underlies it.”
Other experiments mentioned at ECTC targeted on streamlining the bonding course of. Several sought to cut back the annealing temperature wanted to type bonds—sometimes round 300 °C—as to reduce any threat of harm to the chips from the extended heating. Researchers from
Applied Materials introduced progress on a way to radically cut back the time wanted for annealing—from hours to simply 5 minutes.
CoWs That Are Outstanding within the Field
Imec used plasma etching to cube up chips and provides them chamfered corners. The approach relieves mechanical stress that would intrude with bonding.Imec
Chip-on-wafer (CoW) hybrid bonding is extra helpful to makers of superior CPUs and GPUs in the mean time: It permits chipmakers to stack
chiplets of various sizes and to take a look at every chip earlier than it’s sure to one other, making certain that they aren’t dooming an costly CPU with a single flawed half.
But CoW comes with the entire difficulties of WoW and fewer of the choices to alleviate them. For instance, CMP is designed to flatten wafers, not particular person dies. Once dies have been minimize from their supply wafer and examined, there’s much less that may be achieved to enhance their readiness for bonding.
Nevertheless, researchers at
Intel reported CoW hybrid bonds with a 3-μm pitch, and, as talked about, a staff at Imec managed 2 μm, largely by making the transferred dies very flat whereas they have been nonetheless hooked up to the wafer and maintaining them additional clear all through the method. Both teams used plasma etching to cube up the dies as a substitute of the standard technique, which makes use of a specialised blade. Unlike a blade, plasma etching doesn’t lead to chipping on the edges, which creates particles that would intrude with connections. It additionally allowed the Imec group to form the die, making chamfered corners that relieve mechanical stress that would break connections.
CoW hybrid bonding goes to be vital to the way forward for high-bandwidth reminiscence (HBM), in accordance to a number of researchers at ECTC. HBM is a stack of DRAM dies—at present 8 to 12 dies excessive—atop a control-logic chip. Often positioned inside the similar package deal as high-end
GPUs, HBM is essential to dealing with the tsunami of information wanted to run massive language fashions like ChatGPT. Today, HBM dies are stacked utilizing microbump know-how, so there are tiny balls of solder surrounded by an natural filler between every layer.
But with AI pushing reminiscence demand even greater, DRAM makers need to stack 20 layers or extra in HBM chips. The quantity that microbumps take up signifies that these stacks will quickly be too tall to match correctly within the package deal with GPUs. Hybrid bonding would shrink the peak of HBMs and likewise make it simpler to take away extra warmth from the package deal, as a result of there can be much less thermal resistance between its layers.
“I think it’s possible to make a more-than-20-layer stack using this technology.” —Hyeonmin Lee, Samsung
At ECTC, Samsung engineers confirmed that hybrid bonding may yield a 16-layer HBM stack. “I think it’s possible to make a more-than-20-layer stack using this technology,” says
Hyeonmin Lee, a senior engineer at Samsung. Other new CoW know-how may additionally assist convey hybrid bonding to high-bandwidth reminiscence. Researchers at CEA Leti are exploring what’s referred to as self-alignment know-how, says Souriau. That would assist guarantee good CoW connections utilizing simply chemical processes. Some components of every floor can be made hydrophobic and a few hydrophilic, leading to surfaces that will slide into place robotically.
At ECTC, researchers from Tohoku University and Yamaha Robotics reported work on the same scheme, utilizing the floor pressure of water to align 5-μm pads on experimental DRAM chips with higher than 50-nm accuracy.
The Bounds of Hybrid Bonding
Researchers will nearly definitely hold lowering the pitch of hybrid-bonding connections. A 200-nm WoW pitch is not only doable however fascinating,
Han-Jong Chia, a mission supervisor for pathfinding programs at Taiwan Semiconductor Manufacturing Co. , informed engineers at ECTC. Within two years, TSMC plans to introduce a know-how known as bottom energy supply. (Intel plans the identical for the tip of this 12 months.) That’s a know-how that places the chip’s chunky power-delivery interconnects beneath the floor of the silicon as a substitute of above it. With these energy conduits out of the best way, the uppermost ranges can join higher to smaller hybrid-bonding bond pads, TSMC researchers calculate. Backside energy supply with 200-nm bond pads would minimize down the capacitance of 3D connections a lot {that a} measure of vitality effectivity and sign velocity can be as a lot as eight occasions higher than what might be achieved with 400-nm bond pads.
Chip-on-wafer hybrid bonding is extra helpful than wafer-on-wafer bonding, in that it will possibly place dies of 1 measurement onto a wafer of bigger dies. However, the density of connections that may be achieved is decrease than for wafer-on-wafer bonding.Imec
At some level sooner or later, if bond pitches slim even additional, Chia suggests, it’d turn into sensible to “fold” blocks of circuitry so they’re constructed throughout two wafers. That approach a few of what are actually lengthy connections inside the block may give you the option to take a vertical shortcut, doubtlessly rushing computations and reducing energy consumption.
And hybrid bonding might not be restricted to silicon. “Today there is a lot of development in silicon-to-silicon wafers, but we are also looking to do hybrid bonding between gallium nitride and silicon wafers and glass wafers…everything on everything,” says CEA Leti’s Souriau. His group even introduced analysis on hybrid bonding for quantum-computing chips, which includes aligning and bonding superconducting niobium as a substitute of copper.
“It’s difficult to say what the limit will be,” Souriau says. “Things are moving very fast.”
This article was up to date on 11 August 2024.
This article seems within the September 2024 print situation.
From Your Site Articles
Related Articles Around the Web