Practical Winery
58-D Paul Drive, San Rafael, CA 94903-2054
phone:415/479-5819 · fax:415/492-9325
This article is from the July/August 2005 issue of Practical Winery & Vineyard Magazine. Order current or back issues here.



BY Henry Work
Plumpjack Winery (Oakville, CA) offered premium Cabernet Sauvignon in bottles sealed with screwcaps in the late 1990s. Since then, PlumpJack’s continued success has opened a lot of eyes, and the premium wine industry has begun to consider use of screwcaps.

Another winery shifting to screwcaps is Hogue Cellars (Prosser, WA). David Forsyth (director of winemaking) stated in June 2004 (after completing a 2 1/2-year closure study), “Bottling our wines with screwcaps will yield tremendous benefits: elimination of cork taint, consistent and appropriate ageing, lower SO2 levels at bottling, and greater ease in opening bottles.”

Screwcaps, however, are not new. Threaded aluminum ROTE (Roll-on, tamper-evident) or ROPP (Roll-on, pilfer-proof) closures were introduced in the 1960s. Researchers at Dow Chemical in 1933 discovered polyvinylidene chloride, the main material in the liner-seal that is in contact with wine, subsequently branded as “Saran” and “Saranex” (Saran sandwiched between layers of polyethlylene or polyethylene-vinylacetate). When a tin/saran liner is used, it is the tin layer that creates a barrier to oxygen ingress.

Wineries such as E&J Gallo (Modesto, CA) have been producing wines with screwcap in their economy and beverage categories for more than 30 years, notes Mary Wagner, chief technical officer.

“In Finland, Norway, Sweden, and Switzerland, wine under screwcap has been popular for more than 30 years,” reports George Thomson (Crown Cork/United Closures & Plastics [UCP] in the United Kingdom).

With an increasing demand for screwcaps, more suppliers are entering the marketplace. This report will examine quality control factors and standards for the three critical legs in screwcap usage: screwcap suppliers and manufacturers, critical factors in glass bottles, and screwcap application at bottling.

Suppliers / Manufacturers
of screwcaps

Pechiney (now Alcan Packaging), with its Stelvin brand, was one of the first suppliers to offer a long skirt (30mm x 60 mm) screwcap for premium wines. Scott Laboratories offers the Alplast brand. More recent entries are Cork Supply USA with SAVin from Metal Closures Group in South Africa, Oeneo Closures (formerly Sabaté) with the S-Cap, G-3 Enterprises (exclusive U.S. manufacturer of Royal 90™), Richard’s Packaging (representing Crown Cork/UCP), Alcoa, Auscap (Australia), and the Guala Closures Group.

Some of these suppliers are divisions of larger international closure groups, while others contract to have screwcaps made to their specifications. In most cases, the fabrication plants are already making closures for other food and beverage industries and abide by, or utilize, several quality control certifications.

Certifications include ISO-9000 (certified by the International Organization for Standardization), HACCP (Hazard Analysis and Critical Control Point) typically used for food and food product production, American FDA compliance, and European regulations, including those applicable to the use of heavy metals. Ask your supplier for current confirmation of these certifications.

“Closure liners must comply with U.S. FDA CFR 21 reg. 177.1210 and the internal coatings must comply with U.S. FDA CFR 21 reg. 175.300,” explains George Crochiere, Crochiere & Associates, packaging consultants, (Pepperell, MA). “Compliance does not guarantee good flavor, sealing, or barrier properties.”

While certifications and compliance activities are not an iron-clad guarantee against defects, they certainly minimize problems. Many winemakers agree with the assessment of Aron Hess, executive winemaker at Rex Hill, (Newburg, OR), who suggests that the few problems he has experienced have generally not been the fault of the cap manufacturer.

“Wineries should only obtain closures from manufacturers who produce screwcaps under statistical process control,” notes Thomson (Crown Cork/UCP).

Manufacturing of screwcaps
“The manufacturing of aluminum screwcaps starts with the coating or painting process,” says Crochiere. “The flat aluminum sheet has a total of two to five individual coatings applied to the inside and outside. The coatings provide corrosion protection, lubricity for manufacturing and removal torques, improved adhesion for the inks, and decorative color.

“The aluminum is cut into discs and drawn into the closure shape. Depending on the cap and process, this is accomplished in one to three operations; the result is a cap shell with smooth sides.

“If the cap is to have decoration or a logo applied, this is done next with inks and a clear protective varnish. In this process, each cap is handled, registered, and printed individually whereas short screwcaps (used on beer or water) are typically printed during the coating process.

“The last metal operation trims the cap to its specified length, applies the knurling and liner-retaining feature, and makes the cuts for the pilfer-band bridges (breakaway feature).

“The final manufacturing process is insertion of a liner; these are either provided as cut discs or as a roll of material that is cut in the insertion process.

“The screwcap manufacturer conducts quality checks at each stage of this process. Final quality control process should include but not be limited to: closure application, sealing, pilferband operation, removal torques, correct liner material, and coating and ink application, color, placement and adhesion.”

Seal manufacturing processes
and shipping

The actual sealing discs, and their layered materials, are normally made by companies other than the cap manufacturer. The seal material is made by a combination of extrusion, co-extrusion and/or lamination processes. The material is shipped to the cap manufacturer either in sheets, which the cap manufacturer cuts to size as they are inserted into the cap, or as pre-cut discs to save on shipping and scrap costs.

It is important to understand the potential areas for contamination in the supply chain of screwcaps and liners. As noted, the seals are normally manufactured in different plants from where the screwcap is made.

Trucks and containers carry all sorts of products and chemicals (some of them volatile), which could be absorbed by the plastics. Again, as part of the cap manufacturing process, inspection of incoming raw materials should be part of the quality control process. (More on this in the “Bottling line application” section below.)

Sealing material options / selection
“Seals are made to provide a compromise of resilient softness for sealing properties and barrier properties,” says Crochiere, “to minimize oxygen ingress, carbonation loss, loss of volatiles and flavor neutrality; typically these are opposing properties for polymers. Seals are made in three different ways; formed in place PVC, single material discs, and multi-layer discs of resilient cushion layers and barrier material layers.”

“The oxygen transmission rate for Saranex is only slightly greater than tin-saran or aluminum-saran,” says Jim Peck, G-3 Enterprises (Modesto, CA).

“Each sealant material and configuration has its own sealing, flavor, and barrier properties,” adds Crochiere. “Changes in closure and/or liner supply should be carefully evaluated. In addition to the sealant material, the closing conditions also affect the barrier properties as different closing conditions can increase the length and reliability of the seal on the bottle.

“Based on work started in the beer industry in 1982, the sealant industry has developed barrier materials and techniques to evaluate them under commercial conditions for oxygen ingress, carbonation loss, and the ingress of TCA, TeCA, and other undesirable contaminants. The methods are now being used for wine packaging.

“An oxygen permeability coefficient will tell you how many micro-grams of oxygen permeate through a given package component per day. Combined with the volume of the package, an ingress rate in ppb of total package oxygen per day for a closure/liner type, closing condition and bottle size can be determined. These units are the same as those used to measure oxygen levels at filling and can be added together to determine the total oxygen in the package from filling and ingress over time. Versions of this test are also used to evaluate the performance of closures when exposed to abuses like warehouse stacking, transportation, temperature changes, or impact on the closure.”

“Another term for oxygen permeability coefficient is oxygen transmission rate, which would include not only permeation but any micro-leakage,” notes Jim Peck.

“When selecting a new closure and liner, a comparison of oxygen ingress coefficients or rates to existing closure types will help predict the shelf-life and development of the wine with the new closure,” adds Crochiere. “In addition, comparison of TCA or TeCA barrier properties can minimize the risk of cellar or transportation contamination. Typically, the better the barrier properties, the higher the cost.

“A chart of oxygen permeability coefficients for some typical wine closures is in Table I. These numbers have been generated from numerous tests over the past seven years and should only be used as an indication of what is possible.”

The choice between Saran/Tin or Saranex seals generally reflects the winemakers’ decision on whether to seal the bottle completely for a short-lived wine, such as Sauvignon Blanc, Riesling, or Pinot Grigio, or to allow some permeability, for ageing red wines. However, both Patrick Pickett, (winemaker for Pepi, Oakville, CA), and Aron Hess, have observed relatively little difference in the actual taste of wines using either of these two seal types with whites or reds.

“It is a huge debate on how much oxygen,” says Thomson (Crown Cork/UCP). “The technology is available; unfortunately Crown Cork/UCP does not have a fix on the quantities required over time.”

Wineries using screwcaps are trying to get a “fix” on the right amount of oxygen for their various wines. Hogue uses the Saranex seal for both red and white wines, while John Conover (general manager at Plump Jack Winery), reports using the Saran/Tin seal for their reds, and some of their white wines. All of which suggests that the Saran seal for whites and Saranex seal for reds guideline can have many exceptions.

Inspecting screwcaps
Your supplier should have an inspection procedure for all incoming products from its manufacturers. This should provide a second line of defense against possible manufacturing errors. All suppliers contacted for this report did have a receiving inspection procedure in place. These include checks for the cap artwork and colors, that the correct seal-disc has been properly inserted, and that the packaging integrity has been maintained. If you are using a new supplier, it would be worthwhile to ask what checks are included in their receiving inspection.

The distributor’s inspection procedures should also include: inspection of the packaging for damage (especially crushed contents, but also rips and tears); any odd odors; sample inspection of the caps to ensure they conform to size, thickness, skirt length, and style; sampling of the seals to ensure they are the correct type and inserted correctly (since the layers in the saran/tin-type seals are not symmetrical, they must be inserted with the saran side to face the wine); proper printing, and embossing. Colors should match between the top and skirt (they are printed separately), and should match the customer’s specifications. Break points (bridge line) must be properly perforated; and general cleanliness of the caps and their packaging must be assured.

Screwcaps and glass bottles
Not all screwcaps are compatible with all styles of glass bottles, and vice-versa. The bottle finish design and corresponding neck diameter must be compatible with the screwcap and closing equipment. Cap manufacturers and distributors want to inspect a bottle sample the winery proposes to use to make sure their cap will fit properly, or can supply design specifications for a glass manufacturer. Aron Hess emphasizes: “Screwcap manufacturers who do not insist on matching their cap to a specific bottle/finish tend to supply more problematical caps.”

This problem was echoed by another winemaker, Tony Bish (Sacred Hill Wines in Hawke’s Bay, New Zealand), who received a shipment of bottles with a slightly elliptical neck, some outside the range of specifications. The screwcap would not fully descend onto the bottle. The capping head would come down and impact the closure and tighten onto the bottle, resulting in many broken bridges on the cap, compromising the tamper-proofing aspects. The caps did seal correctly, but the result was unsatisfactory due to sharp edges where bridges were broken and unsightly appearance.

A different brand of screwcap worked better, because the caps had a slightly wider skirt to allow more caps to fully seat on the bottle-neck before the capping head was applied. Also, the inside surface of the caps was smoother and the reduced friction allowed caps to seat better. However, since the bottles were out-of-round due to forming irregularities, even this did not fully correct the problem.

The Sacred Hill problem was only solved by disgorging some bottles and re-bottling in new bottles within specifications. Given the tight tolerances required for a proper seal and tamper-proofing, all screwcap suppliers and winemakers recommend that great effort be taken to ensure that the screwcap and glass bottle styles be absolutely compatible.

“Richards Packaging has had several experiences with matching of bottle and closure,” reports Bob Boord. “The Glass Packaging Institute (GPI) provides neck finish specifications, but many small deviations have crept into bottle neck finishes over the years. In addition, molds wear. The result is that constant attention needs to be given to the proper matching up of neck finishes on bottles with the closure specifications.

“In one situation, a customer used seven different glass bottles. Analysis found that there were seven variations of the GPI neck finish, and some of those deviations were serious. In another situation, a glass manufacturer built new neck finish inserts, but the new specifications changed the fit of the closure on the bottle.

“Manufacturers of bottles and closures do not communicate well with each other, so it becomes important for the winery-buyer and the distributor salesperson to work together to provide bottle samples to the closure manufacturer for testing to assure good fit. Whenever a new bottle is introduced or existing bottle molds are rebuilt or modified, samples of all cavities should be sent to the screwcap manufacturer for analysis.”

The actual sealing of the screwcap is achieved at the top of the outside rim radius of the glass bottle. Therefore, this rim must be smooth and correctly curved. The glass thread must start a specified distance below this rim to accommodate the seal.

“The holding power of the cap is achieved by the threads, while the pilfer-band locks the cap in place,” explains Crochiere. In addition to the proper surfacing at the top edge, the thread and pilfer band ridge below the thread tolerances are relatively tight, thus the need for coordinating the bottle and cap specifications.

Bottling line application
Prior to and at bottling, a number of inspections and checks must be made. Most of the caps are packaged bulk in plastic bags and shipped in cardboard boxes. The boxes must not show any evidence of being crushed, which could dent or deform the caps; nor should the plastic be ripped or torn, in order to keep the caps clean and protect them from odors. Clean, dry storage in the winery is essential.

“Bottles should be inspected for straight and concentric necks in addition to dimensional checks to insure that the sealing area, threads, pilfer step and neck comply with specifications,” says Crochiere. “Bottles should also be inspected for manufacturing imperfections or damage (glass flash or slivers at the mold-parting lines) that could result in unacceptable performance of the screwcap sealing system. Don’t forget to check for cleanliness and off-odors in the bottles and cases.”

Alignment of the capper is absolutely critical. Kristi Koford (winemaker at Napa Wine Company, Oakville, CA), notes that one screwcap manufacturer insists on inspecting the head set-up for first-time users. She appreciates this service since they have previously experienced a brand that wouldn’t seat on the bottles and then jammed the capping machinery.

The capper is a complex piece of machinery with a precisely-shaped head (pressure block) and several sets of rollers designed to mold the screwcap in a specific manner. Precision springs apply a pressure of around 400 lbs. force (28.12kgf/cm2) to the pressure block. This squeezes and forms the screwcap on the bottle’s top sealing edge, forming the seal. This part of the sealing is referred to as a reform or redraw. The capper should be set to the particular screwcap manufacturer’s settings.

Wineries should consult with Alcoa, Zalkin, Silgan, Arol, Bertolaso or other capper manufacturers to determine top pressure in a capper.

“There are three general types/ shapes of pressure blocks,” explains Crochiere: “1) a flat surface that produces a top seal only, 2) a partial or slight curved block that produces a slight corner seal, and 3) a full reform that draws or wraps the closure and liner around the top of the glass and seals 1.5 to 2.0 mm down the outside of the bottle.”
The rollers that form the threads need to be exactly aligned and moving at a speed that will allow them to be inserted and released at the proper time.

Steve Rasmussen (owner of S.L.O. Bottling Services), notes that the capper head needs to move slow enough in order to allow these rollers to accurately form the threads. If speed is essential, more heads should be installed.

The second set of rollers simultaneously crimps (undertucks) the aluminum closure under the pilfer ring and smooths the skirt to complete the seal and hold the cap in place. The head and screwcap manufacturers recommend specific pressures for each of these sets of rollers.

Another important factor is the head space to allow for wine expansion without creating excessive internal pressure in the sealed bottle. Generally, screwcap manufacturers suggest around 2% of the wine volume. This fill level should be checked periodically throughout the bottling process.

As more winemakers become familiar with using screwcaps, one current debate is on sparging with either nitrogen or CO2 prior to cap placement, to displace any oxygen. Some winemakers and bottling line companies use nitrogen gas, while others are using liquid nitrogen. Patrick Pickett (Pepi Wines), uses liquid nitrogen that requires little time to displace the oxygen.

“The total packaging oxygen (TPO) exposure to the wine is the sum of the oxygen pickup during transfer to the filler,” explains Crochiere, “pickup as the wine is being filled into the bottle, the oxygen added from the head space, and the oxygen ingress through the package (closure) during the life of the product. Typically the two largest contributors are the head space and the ingress.”

Figure I illustrates the amount of oxygen added to the filled product as a function of volume of head space. The chart shows two relationships, one with a complete flush as with liquid nitrogen, the other with no flush.

“If you use inert gas,” adds Crochiere, “the head space flush will not be as thorough due to air turbulence and allowed time. When using liquid nitrogen, there should be excess added so that bottles are sealed with 10 to 15 psi nitrogen pressure.
“Selection of the sealing liner and optimum reform will minimize ingress as illustrated by the oxygen permeability coefficients and ingress rates in Table I. The use of inert gas to flush the empty bottle before filling (with or without alternating vacuum) would optimize the filling line for minimum oxygen pickup.”

David Wright, production manager for Pacific Wine Partners (Gonzales, CA), uses CO2 which, being heavier, displaces oxygen. Hogue Cellar’s quality supervisor, Mei Yeung, notes that because screwcaps allow for relatively large head space, when compared to cork closures, the use of nitrogen, carbon dioxide, or argon is necessary to remove the oxygen from that space. Hogue uses liquid nitrogen. On the bottling line, the gas (regardless of type), is delivered just after filling and before cap application.

Quality control of the caps does vary. Jose Mendez, the bottling line supervisor for Ultima Mobile Bottling, has, on occasion, seen a difference in skirt lengths of 1 to 2 millimeters. The longer ones can block the cap feed. He also watches for the seal discs falling out of caps as they move through the cap feeding system.

After bottling, case storage is recommended at no more than three pallets high, in order to prevent excessive compression on the caps. PlumpJack Winery solves the problem by shipping bottles horizontally in six-bottle wooden boxes. Hogue Cellars stores screwcapped wine without pallets four-cases high for even distribution of vertical forces.

Testing on bottling line
As a basic method to ensure that bottles are properly sealed, most suppliers recommend periodic testing of the torque required to remove a screwcap. Several supply a torque tester to accurately and easily perform this test. Frequency of this testing is up to the winery, although one bottle, every one to two hours per capper being used, is a commonly mentioned standard.

Wright suggests that the torque specifications should be in the range of 5 to 15 inch-pounds. The test is conducted with two twists. The initial twist breaks the seal and should be on the higher side of the range. The second breaks the aluminum “bridges” connecting the threaded portion to the lower skirt, and should be no less than the lower number in the range. (Normally the wine consumer would unseal and break the “bridges” with one motion.)

A pressure test is used by Wright to test the seal. This consists of injecting CO2, to apply pressure into the bottle via a threaded needle, while the bottle is under water. As the pressure is increased, an operator watches for bubbles escaping from under the cap. Normally the cap should hold a minimum of 20 lbs/in2 of pressure for still wines. Wright emphasizes, “This is an important test as the internal pressure in the head space of a bottle can vary drastically due to elevation and temperature changes.”

“The Secure Seal Test (pressure test) injects any pressurized gas at a controlled rate into the bottle via a threaded piercing needle,” says Crochiere. “The specification of 150 psi is for carbonated soft drinks, beers should hold about 100 psi, still products like wine only need a closure that holds 40 to 50 psi. This is the force required to compress the liner material and allow the air to vent. A closure that is applied incorrectly can leak at very low pressures.”

Several companies supply testing equipment for these checks. Contact your screwcap supplier or these testing equipment suppliers: Secure Pak or Dillon.

Additionally, a bottling line worker should make a visual inspection of the caps; noting the smoothness of the threads, any deformation or bulges at the top, and a clean crimp line. When the top of the cap is removed, it should separate cleanly at the “bridges,” with no torn or sharp edges to cut the opener’s fingers.

“Inspection should also include the coatings and inks — the more decorated the cap, the more likely there will be paint and ink dust generated in the screwcap hopper,” adds Crochiere. “This can be unsightly, but more important, the dust will collect inside the caps and contaminate the wine. In addition, check for coating lifting at the cut edges of the pilfer bridges and the bottom of the cap. Dark color caps show scratches much more than light color caps as the exposed aluminum makes a sharp contrast.”

QC issues for further research
In 1999, Peter Godden and colleagues at the Australian Wine Research Institute (AWRI) undertook an extensive research project on closures for wine bottles — including the use of corks, synthetic corks, and screwcaps. One result of their testing was that screwcaps showed the least reduction in free SO2 of all closures tested.2

Having noted a similar maintenance of SO2 levels by screwcaps, several winemakers interviewed for this article including Shaun Richardson, vice president and winemaker at Clos Pegase (Calistoga, CA), are looking into the possibility of reducing the SO2 level at bottling.

However, the AWRI closure research project also indicated, through sensory testing, that the screwcap closure had the highest “reduced” aroma. It was described as “struck flint” or rubbery.1 The authors reporting the study subsequently note that, after 36 months in the bottle, the same wine sealed with the corks also exhibited similar aromas. Further exploration of the causes and/or relation between the SO2 levels and reduced aromas issues will be required.

Perhaps associated with this reduction are the yet-to-be totally defined differences in permeability of sealing materials in discs of various screwcap brands. While Conover indicated that he does not want breathability with the PlumpJack Cabernets and hence uses the Saran/Tin seal, Richardson uses the Saran/Tin for the Clos Pegase Sauvignon Blanc wines, and Saranex disc seals for the Chardonnay, in order to allow some oxygenation during ageing. The jury appears to still be out on the issues of seals and wine type, with more experimentation and experience necessary.

Screwcaps have a history of use in many industries. They are relatively new as a premium wine closure, but have a long history of use with low-priced wines. The technology and materials are well-tested and appear to be suitable for wine. If winemakers make sure their caps and glass are compatible, are cautious with new suppliers, and test frequently with the first bottlings, screwcaps should have a positive impact on wine, by providing, as Patrick Pickett notes, “consistency bottle to bottle.”

  1. Francis, L., J. Field, M. Gishen, A. Coulter, P. Valente, K. Lattey, P. Høj, E. Robinson, P. Godden. (2003, August). “The AWRI closure trial: sensory evaluation data 36 months after bottling.” Australian & New Zealand Grapegrower & Winemaker, 475, 59–64.
  2. Godden, P., L. Francis, J. Field, M. Gishen, A. Coulter, P. Valente, P. Høj, E. Robinson. (2001, September). “Results of an AWRI trial investigating the technical performance of various types of wine closure: Part II – Wine composition up to 20 months post-bottling.” Australian Grapegrower & Winemaker, 452, 89–98