MECHATRONICS: How Solar Panels are Made

Friday, 17 July 2015

How Solar Panels are Made

Crystal growing

SolarWorld heats and melts polysilicon rock until it forms a whitehot liquid, then re-fuses the molten silicon into a single giant crystal in which all atoms are perfectly aligned in a desired structure and orientation.

Charging

The magic starts with about 250 pounds of polysilicon rocks carefully stacked in a quartz crucible. The only other ingredient is a silicon disk impregnated with a tiny amount of boron. The addition of the oron dopant ensures that the resulting crystal will bear a positive potential electrical orientation. The crucible is encased within thick walls of insulating graphite and locked inside a cylindrical furnace.

Melting

As the crystal-growing furnace heats up to temperatures ranging around 2,500 degrees Fahrenheit, its silicon contents melt into a shimmering slurry. Once computerized monitors register the right temperature and atmospheric conditions, the alchemy begins. A silicon seed crystal, hung from a narrow cable attached to a rotary device atop the furnace, is slowly lowered into the melt.

Growing

The crucible starts to turn, and the seed crystal begins to rotate in the opposite direction. The silicon melt freezes onto the seed crystal, matching the seed’s crystalline structure. The crystal grows, the cable and seed slowly ascend, and the crystal elongates at a controlled width. As the growth depletes the silicon slurry, the crucible also rises.

Cooling

Flash forward about 2.5 days since the crucible was charged with polysilicon: After hours of cooling to about 300 degrees Fahrenheit, the furnace hood and shaft lift away from the crucible encasement, slowly swing to one side and reveal a completed cylindrical crystal, ready to move to the second step and next production room.

A silicon crystal must change shape several times before it winds up as the precisely calibrated wafers that form the foundations of photovoltaic cells.

Cutting

First, a saw cuts off the crystal’s so-called top and tail, so that a crystal of uniform width remains. Typically, wafering saws draw thin wire bearing a liquid abrasive across the crystal’s surface. (Below, a machine mounted with a giant donutlike steel blade does the cutting.) Wire saws also cut the crystal into ingots measuring 2 feet or less. Steel holders are mounted on the ends of these ingots for the next step.

Squaring

Mounted ingots are placed standing on end in a rack bearing 16 at a time inside another wireslicing machine. There, wire running in a lattice configuration descends on the ingots to shear off four rounded segments, leaving flat sides. The result: The ingots now have a square crosssection, except for still-rounded corners.

Slicing

The next wire saw is more intricate yet. A wire winding hundreds of times between two cylindrical drums forms a web of parallel, tightly spaced segments. As the wire unspools through the machine, ingots mounted sideways on glass and metal holders are pressed two at a time through the wire web, slicing them into the thickness of slim business cards. Each millimeter of crystal yields about 21/2 wafers. Detached from their holders, the wafers are loaded into carriers, or boats, for transport to the next step.

Converting wafers into cells

At this point, a wafer is no more capable of producing electricity than a sliver of river rock. The wafer is the main building block of a PV cell, but so far its only notable characteristics are its crystal structure and positive potential orientation. All of that changes in the third, multistep, cell-production phase of PV manufacturing.

Etching

In the only phase requiring a designated clean room, a series of intricate chemical and heat treatments converts the blank, grey wafers into productive, blue cells. A so-called texture etch, for instance, removes a tiny layer of silicon, relying on the underlying crystal structure to reveal an irregular pattern of pyramids. The surface of pyramids – so small they’re invisible to the naked eye – absorbs more light.

Diffusing

Next, wafers are moved in cartridges into long, cylindrical, ovenlike chambers in which phosphorus is diffused into a thin layer of the wafer surface. The molecular-level impregnation occurs as the wafer surface is exposed to phosphorus gas at a high heat, a step that gives the surface a negative potential electrical orientation. The combination of that layer and the boron-doped layer below creates a positive-negative, or P/N, junction – a critical partition in the functioning of a PV cell.

Coloring and Printing

The burgeoning, still-grey cells move in trays into heavy vacuum chambers where blue-purple silicon nitride is deposited onto their tops. The coating with silicon nitride – yet another member of the silicon family of materials – is designed to reduce reflection even further in the energy-dense blue end of the light spectrum. It leaves the cells with their final, dark color. Now, the cells can optimally gather photons and produce electricity. They lack, however, any mechanism to collect and forward the power. So, in a series of silkscreen- like steps, metals are printed on both sides of the cell, adding pin-stripe "fingers" and bus-bar circuitry. A functioning cell is born – only sunshine needed.

Stringing cells into solar panels

Each phase of production depends on processes with flavors all their own. Careful control of heating and cooling dominates crystal growing. Wafering employs abrasion and cutting. Cell production concentrates on chemistry. Any factory process would be incomplete without a final assembly step, and in PV such a step is known as moduling.

Soldering

At SolarWorld, module manufacturing is a highly automated process, relying on robust steel robotics to undertake the increasingly heavy lifting of assembling lightweight PV cells into modules weighing around 45 pounds apiece. Each robotic tool works within a safety fence that, by design, excludes people. First, cells are soldered together into strings of 10, using an over-under-over-under pattern of metal connectors to link the cells. Six strings are laid out to form a rectangular matrix of 60 cells. Each matrix is laminated onto glass.

Framing

To become a module, however, each laminate requires not only a frame to provide protection against weather and other impacts but also a junction box to enable connections among modules or with an inverter-bound conduit. Robots affix those, too.

Inspection and Shipping