Why Serious Collectors Are Swapping Century-Old Corks for Screw Caps

Cork at the table and what happens in the cellar

Natural cork still signals luxury in most dining rooms. I have opened thousands of bottles on the floor, and guests still notice the pop. What changed in the last twenty years is what happens after that moment: whether the wine in the glass matches the label on the neck, and whether the other eleven bottles in the case will taste the same in five years.

Roughly 2.6% of wines closed with natural cork showed TCA or another moldy contaminant in one critic’s 2023 review set of more than 1,250 bottles (Sullivan, 2024). Reviews of global production still put 2–5% of bottled wine in the “corked” bucket, with industry losses above a billion dollars a year (Zhou et al., 2024). That matters when you buy by the case. Add oxygen transmission rate (OTR)—how much oxygen passes through the seal after bottling—and natural cork spans a range wide enough that Keller, founder of the VinPerfect screw-cap program, argues the “average” cork misleads buyers because OTR can vary across two orders of magnitude (Miller, 2025).

Modern screw caps use engineered liners—often Saranex or Saran/Tin layers inside the metal shell—that let a winery target a chosen OTR instead of inheriting whatever each tree ring allows. The swap serious drinkers are making is less about auction catalogs, which remain dominated by cork-sealed Bordeaux and Burgundy, and more about drinking what you paid for: fewer spoiled bottles and less variation inside a case. The sections below cover taint, oxygen, liner choice, who is actually changing closures, and what to do in your cellar.

Cork taint: what “3%” still means in 2026

Cork taint usually means TCA (2,4,6-trichloroanisole), a mold-related compound that strips fruit and leaves wet cardboard, damp basement, or musty notes. Thresholds are tiny—often cited around 2–5 nanograms per liter—so a clean wine can be ruined by trace contamination from cork or, less often, from cellar air and certain other closures (Zhou et al., 2024).

The old rule of thumb was about 3% of corks tainted. Screening has improved. The International Wine Challenge fault database, covering 117,732 samples from 2007–2017, reported cork taint at about 1.3% for cork-sealed wines (Godden, 2020). UC Davis trade reporting has cited 1–2% after better cork analysis (Miller, 2025). Yet in Pacific Northwest professional tastings, 2.6% of natural-cork bottles still failed for TCA or other moldy faults in 2023, close to 3.05% in 2021 (Sullivan, 2024). Progress is real at the industry level—tainted bottles in that critic’s full sample fell from 6.8% in 2015 to 1.1% in 2023—but much of that drop came from switching to DIAM and other technical closures, not from whole cork becoming faultless (Sullivan, 2024).

At the table, the tasting pour still exists to catch faults like these before the bottle is accepted for the table—a point I covered in an earlier piece on navigating wine lists. If you smell TCA, say so clearly. A professional team will re-taste with you. Preference alone does not reopen a sound bottle.

Oxygen ingress: what collectors need to know

Aging is partly chemistry and partly oxygen budgeting. At bottling, dissolved oxygen and headspace gas burn through free sulfur dioxide (SO₂), wine’s main preservative. After that, OTR governs how fast the wine moves toward oxidation—browning, dried fruit, loss of freshness—or, with too little oxygen, toward reduced aromas like flint or rubber.

Andrew Waterhouse at UC Davis puts the benchmark plainly: for natural cork, about one milligram of oxygen per year passes through a new cork on average, and the market has spent decades trying to mimic that curve with other closures (Miller, 2025). The problem is the word average. Godden’s summary of more than twenty years of Australian Wine Research Institute (AWRI) work shows screw caps with the lowest OTR and very low variability bottle to bottle, while natural cork shows higher OTR spread, which shows up as inconsistent bottles in the same case (Godden, 2020).

The 1999 AWRI closure trial bottled a high-quality Semillon under 14 closure types—two natural cork grades, two technical corks, nine synthetics, and one screw cap—about 8,600 bottles total, stored inverted at 15 °C (Godden, 2020). Through 20 months, the screw cap held the most SO₂, kept fruit freshest, and browned slowest; technical corks followed, then natural cork (Godden et al., 2001). Laboratory testing stopped after three years because results were clear-cut, but photographs at two, five, and ten years kept the same order: screw cap and the Altec technical cork (forerunner of DIAM) ahead of natural cork (Godden, 2020).

In a 2018 presentation set, the AWRI compared reds and whites sealed both ways. Four of 30 natural-cork bottles showed cork-related taint; none of six Diam bottles did. One bottle each of three reds under natural cork was markedly oxidized while siblings under cork were fine; screw-capped bottles of the same wines matched the best cork examples (Godden, 2020). Same vintage, same closure class, three different outcomes—that is the practical problem for anyone cellaring by the case.

Screw caps are not one product: liners and trade-offs

Not every metal cap behaves the same. The liner touching the glass rim sets OTR. Saran Tin liners aim for near-zero oxygen ingress; Saranex liners allow a controlled, cork-like exchange (Miller, 2025). Winemakers pick the curve; the cap holds the liner in place.

A controlled industrial study of a Merlot–Tannat red bottled at 1.5 mg total oxygen per bottle tested eight closures with manufacturer OTR values spanning roughly 5 to 67 μg per day (Valentin et al., 2017). Low-OTR technical corks and Saranex screw caps held more free SO₂ and less color drift from months four through eighteen; high-OTR synthetics and some screw caps without PVDC aged faster on paper. Sensory panels at 10 and 17 months did not always rank those chemical gaps the way lab numbers suggested—worth remembering before you treat OTR charts as a guarantee in the glass (Valentin et al., 2017).

Other trials complicate the picture. A 30-month Napa Sauvignon Blanc study with 200 bottles per closure type found higher browning (A420 proxy) under screw cap than under natural cork by the end—about 0.83 versus 0.78 on that scale—while synthetics tracked closer to cork (Crochiara et al., 2022). A five-year Riesling and wooded Chardonnay study reported reduced or struck-flint notes under screw cap and glass ampoules, absent in the natural-cork lots in that trial (Skouroumounis et al., 2005). Godden’s AWRI work stresses the same point: low OTR does not create reduction by itself, but wine with a tendency toward sulfur off-notes will show them more clearly under tight seals (Godden, 2020). Wineries that bottle under Saran Tin, common in Australia and New Zealand, manage dissolved oxygen and SO₂ at filling accordingly (Miller, 2025).

Who is actually “swapping,” and what changed since Clare Valley

The headline speaks to collectors, but the data split by channel. James Halliday’s Australian Wine Companion submissions ran 51.5% screw cap in 2007 and about 90% by 2016–2018, with 97.3% of white entries under screw cap in 2018 (Godden, 2020). New Zealand and Clare Valley Riesling producers moved in large numbers after the AWRI trials—not because metal caps were cheap, but because consistency mattered more for aromatic whites meant to age.

In the U.S., a 2021 winery survey cited in trade press still showed 70% preferring cork, but screw-cap use among respondents had jumped toward 52%, up from roughly 30–40% in prior years (Miller, 2025). On my floor, guests under forty rarely object to a screw cap on a $40 Chinon or Oregon Pinot; guests who built cellars in the 1980s sometimes do, until they taste the wine.

Auction rooms tell a different story. Christie’s 2025 sales of major cellars—Koch, Schwarz, and similar—moved tens of millions in cork-sealed Bordeaux, Burgundy, and Napa. Hammer price follows provenance, producer, and vintage, not closure type on a weeknight Chenin. The serious collector who is “swapping” is often buying two tracks: investment lots still under cork, and drinking stock under screw cap or DIAM where spoilage and bottle variation would defeat the purpose of the purchase.

Pacific Northwest review data shows the middle ground. Natural cork’s share in one critic’s opens fell from 77% in 2016 to 42.8% in 2023, while DIAM and relatives rose from 5.7% to nearly 40%; screw caps held near 13% (Sullivan, 2024). Case production still skews natural cork; the shift is clear. I cellar both. I reach for screw cap when I want the 2016 Riesling to taste like the 2016 Riesling, not a random bottle from the same case.

What to do at the shop and in the cellar

Match closure to your window. Drink-now Sauvignon or Grüner under a tight screw cap rewards freshness. Structured Cabernet you plan to hold ten years may still arrive under cork or Saranex cap depending on the house—read the tech sheet if the importer posts one.

For natural cork: Store horizontally when aging more than a year or two so the seal stays wet; target 55–59 °F and humidity near 60–70%. Note ullage and provenance if you might sell.

For screw cap: Orientation matters less; temperature swings still cook wine. Re-seal and refrigerate leftovers without guilt.

At the restaurant: Reject clear TCA, oxidation, or heat damage on the taste—not “I wanted more oak.” Use direct language: you believe you detect cork taint and you would like the sommelier to confirm.

When resale matters: Keep invoices, storage logs, and photos. Auction buyers discount mystery cellars regardless of whether the seal was bark or metal.

Do not assume screw cap equals plonk. Do not assume cork equals greatness. The liner and bottling oxygen shaped the bottle as much as the vineyard.

Cork will keep its place in fine dining because the ritual still matters to many guests. For the collector who actually drinks the cellar, the numbers have shifted: a few percent spoiled bottles, wide OTR scatter under bark, and engineered liners that let winemakers choose oxygen the way they choose yeast. Auction Bordeaux will stay corked for years; your Tuesday Barolo and your aged Marlborough Riesling may already sit under metal—and taste better for it.

I have used this line before, aimed at the cellar rather than the guest list: drink the wines you bought to enjoy while they are still the wines you paid for. The best seal is the one that gets you there reliably, even when it does not make a sound.

References

Australian Wine Research Institute. (2020). Summary of AWRI closure trials and other investigations into closure performance (Technical Review No. 248). https://www.awri.com.au/files/attachment/248-october-2020-technical-review-godden/

Crochiara, E., Waterhouse, A. L., & Block, D. E. (2022). Wine closure performance of three common closure types: Chemical and sensory impact on a Sauvignon Blanc wine. Molecules, 27(18), 5881. https://doi.org/10.3390/molecules27185881

Godden, P., Francis, I. L., Field, J., Gishen, M., Coulter, A., Valente, P., Høj, P. B., & Robinson, E. (2001). Wine bottle closures: Physical characteristics and effect on composition and sensory properties of a Semillon wine 1. Performance up to 20 months post-bottling. Australian Journal of Grape and Wine Research, 7(2), 64–105. https://doi.org/10.1111/j.1755-0238.2001.tb00196.x

Miller, C. A. (2025, June 18). The science of aging wine under screw cap. SevenFifty Daily. https://daily.sevenfifty.com/the-science-of-aging-wine-under-screw-cap/

Skouroumounis, G. K., Kwiatkowski, M. J., Francis, I. L., Oakey, H., Capone, D. L., Peng, Z., Duncan, B., Sefton, M. A., & Waters, E. J. (2005). The impact of closure type and storage conditions on the composition, colour and flavour properties of a Riesling and a wooded Chardonnay wine during five years' storage. Australian Journal of Grape and Wine Research, 11(3), 355–368. https://doi.org/10.1111/j.1755-0238.2005.tb00036.x

Sullivan, S. P. (2024, May 22). Use of DIAM and related closures surge, cork taint remains stubborn. Northwest Wine Report. https://www.northwestwinereport.com/2024/05/use-of-diam-surges-cork-taint-remains-stubborn.html

Valentin, E., Lurton, L., & Guinard, A. (2017). Comparison of the effect of 8 closures in controlled industrial conditions on the shelf life of a red wine. BIO Web of Conferences, 9, 02024. https://doi.org/10.1051/bioconf/20170902024

Zhou, H., Xie, Y., Wu, T., Wang, X., Gao, J., Tian, B., Huang, W., You, Y., & Zhan, J. (2024). Cork taint of wines: The formation, analysis, and control of 2,4,6-trichloroanisole. Food Innovation and Advances, 3(2), 111–125. https://doi.org/10.48130/fia-0024-0011