Nonwoven Tote Bags from China: GSM Tolerance, Handle Welds and the rPET Claim That Fails an Audit

A buyer in Rotterdam asked me a question over lunch in March that I did not have a good answer for, and it has been rattling around since. He buys about 300,000 nonwoven tote bags a year, all from one factory near Wenzhou, and he asked: "How much heavier would my bags have to be to actually survive eight trips to the supermarket instead of three?"

The honest answer is that gram weight is almost irrelevant. What kills a nonwoven bag is the heat-welded seam at the handle root, and the polymer it is made from, and nobody ever quotes either of those.

Spunbond nonwoven: what the GSM number really describes

Nonwoven tote bags in the bulk promotional market are made from spunbond polypropylene, or sometimes a spunbond-meltblown-spunbond three-layer construction if the bag is meant to be printed on both sides with high colour density. The weight is quoted as grams per square metre. A cheap promotional bag sits at 60 to 70 gsm. A supermarket bag with a little ambition sits at 80 to 100 gsm. A proper reusable bag for a chain that has a sustainability claim to defend sits at 110 to 130 gsm.

Here is the problem. GSM is measured on the roll, and the roll is inspected at the edges where a spunbond web is usually slightly thicker. A 90 gsm spec delivered as 82 gsm on the average and 76 gsm in the centre of the roll is common, and within many suppliers' reading of "tolerance". Cut a 10 by 10 cm square from the middle of three bags from three different cases and weigh them on a kitchen scale that resolves to 0.1 gram. Multiply by 100. Do it before ordering and again on the first bulk shipment. I have caught a 15 percent shortfall on the second shipment of a four-shipment order, and the only reason I caught it was because the count was light.

The handle is where the bag fails, and the handle is decided by the weld

Look at where the handle meets the body. There are three constructions in the market.

The cheapest is a die-cut handle, where the handle is simply cut out of the bag body itself. It tears from the cut edge inward, usually within ten uses, and the failure is always at the same corner. It is fine for a giveaway at a trade show. It is not a reusable bag.

The middle option is the welded loop handle, where a folded strip of the same nonwoven is ultrasonically or thermally welded onto the body. The weld is the whole bag. A good weld at 100 gsm, with a weld width of 10 mm and a horn that is actually tuned, will hold 15 to 20 kg on a static hang test. A weak weld lets go at 6 kg, and it lets go in the store in front of the customer, which is the worst possible place.

The better option is a handle that is stitched through the body with a reinforcing patch, or a bag with a full-length sewn seam. These need a sewing operation, so they cost maybe 25 to 35 percent more, but they do not delaminate under load.

My acceptance test is a simple load-and-hang: fill the bag with 12 kg of rice or sand, hang it from one handle, wait an hour. Then hang from the other handle, wait an hour. Then swing it ten times and check the weld. A supplier who cannot pass this at their quoted weight either needs a wider weld, a heavier web, or more horn energy, and all three of those are things they will discover in ten minutes if you ask them to prove it before shipping.

RPET, recycled content and the claim that has to survive an audit

Recycled content is now a sales argument in Europe and increasingly in the US, and the documentation chain is where most Chinese nonwoven suppliers are weakest. There are three levels of claim.

The weakest is "eco-friendly material" with no traceability. Ignore it.

The middle is post-industrial recycled polypropylene, which is usually factory scrap reground and re-spun. It is real recycling, it is cheap, and it can be documented with internal records. It also usually means a web with slightly lower tensile strength and more colour variation, so printed bags on rPET often show a faint grey cast under white ink.

The strongest and the one that survives a customer audit is post-consumer rPET certified under GRS, the Global Recycled Standard, with a transaction certificate covering the specific shipment. GRS requires a certified supply chain at each stage and a transaction certificate issued for the actual delivery, not a general company certificate. If a bag supplier shows you a company GRS certificate but cannot produce a transaction certificate for your order, the bag you receive may or may not contain recycled content, but the claim will not survive scrutiny.

Ask for the transaction certificate before the deposit, not after. Suppliers who can genuinely do it will have the paperwork flow already set up with their fabric mill.

Lamination, coating and the smell test

Bags that need photographic printing or a gloss finish get a laminated film on top of the nonwoven. The film, usually a BOPP layer, adds stiffness and print quality and can make a 70 gsm bag feel like it is 110 gsm. It also changes the bag's end-of-life story, because a laminated nonwoven cannot be easily recycled as monomaterial PP, which undercuts the environmental argument the bag was bought to make.

The other issue is smell. Cheap lamination adhesive and some recycled resins leave a distinct odour that appears a day after unpacking. If you are shipping to a retail chain, a musty bag is a return. Ask for a retained sample from the last three production runs and smell them at 24 hours and at one week. Any experienced supplier will have samples on the shelf. Any supplier who says "our bags have no smell" without samples is telling you they have not been asked before.

What I put on the specification sheet now

Construction and weight, with the tolerance stated as an average across the bag rather than a roll spec. Handle type and weld width in millimetres. A stated load test with the pass criterion. Sewn or welded side seams named explicitly, with seam width. Print method, because screen printing, heat transfer and digital give completely different edge definition and abrasion resistance. Recycled content level and the specific certificate type required. And a packaging instruction that says bags per polybag, polybags per carton, and carton compression strength, because a pallet of 300,000 bags is heavy and stacks badly.

None of that is exotic. It is the difference between a bag that shows up on a shelf and a bag that comes back to a customer service desk.