Bakken Update: Mega-Fracs Could Mean CARBO Ceramics Is A Dead Man Walking

| About: CARBO Ceramics (CRR)


Ceramic proppant demand is down 49% year over year.

Operators continue to increase proppant per well, but are choosing to use more frac sand.

A large number of mega-fracs use little to no ceramic proppant.

The current ceramic proppant market is oversupplied and higher oil prices probably will not right the industry.

The drop in oil prices continues to stress all aspects of the US unconventional oil and gas industry. This includes E&Ps, oil service, and midstream. Oil service lowered prices to compete, while midstreams got a smaller cut per barrel. E&Ps had to lower while increasing production per foot. Many focus on production numbers, but lateral length is variable. This can range from 5,000 to 15,000 feet. As a general rule, the longer the lateral, the better the production. If we compare a 5,000-foot lateral producing 100,000 BO in the first year to a 10,000 foot lateral producing 150,000 BO. The shorter lateral produces 20 bbls/ft while the two mile produces only 15 bbls/ft. Operators who drill longer laterals are not necessarily producing better, just drilling more. This increases costs, but in a lower oil price environment, operators are able to hide this with other cost savings.

Click to enlarge

(Source: Greenlight "Motherfracker")

The difference of a 30-year well life versus 50 years provided an entirely different outlook on Pioneer's (NYSE:PXD) results. A longer well life means initial production will decrease, and although the resource garnered may be the same, it takes 50 years for the well to exhaust production. This probably means the well won't produce for the entire time as it will become a stripper well sooner. This brings about another key point that many miss when deciding how good a well is. One of the most important words an oilman will ever use is payback. It refers to how long it takes for a well to pay back the costs associated with getting it turned to sales. This includes many things over and above D&C. Production taxes, differentials, and other costs offset revenues. Once the well reaches payback, every dollar produced is a dollar used towards drilling another well. So locations can become cash cows in the right environment and produce significant income. This is the beauty of horizontal production. Fracking produces large volumes of resource and dollars early. EURs do not provide an understanding of well economics.

Lower oil prices are going to cause a number of E&Ps to default. Marginal producers with high costs are in trouble, but so are service providers. Operators with core acreage will spend most of its time drilling big wells, hoping prices will rise so it can then start drilling marginal leasehold. Core wells can produce 2, 3 or even 5 times the resource of less appealing areas. E&Ps can drill a fraction of the locations and still increase production year over year. Since fewer wells are drilled, all industries are affected. Some names will benefit by holding on and getting more business later. Others will go away or be swallowed up by the competition.

Now that operators are drilling Mega-Fracs, larger volumes of fluids and proppant are used per foot. Some think this may save associated industries, but some do not benefit. We like some frac fluids providers, especially Flotek (NYSE:FTK). Slickwater well designs encourage the use of frac fluids. Now that water is much cheaper and many are hooked to pipe, slickwater isn't as costly. Proppant is another important variable of well design. Proppant comes in many shapes, sizes, and strengths. As the name indicates, proppant "props" open the induced fractures created during completions work. To provide an idea of how this works, we will go over a remedial description of a horizontal well. Initially, the well is drilled. After drilling down approximately 10,000 feet, the drill moves from vertical to horizontal. There are many reasons for this, but the most significant is getting as much of the source rock or interval in contact with the well. If the geology provides shale that is 100 feet thick, that is all a vertical will be able to frac and produce. This makes areas like the Bakken uneconomic using vertical production, even at very high oil prices. But if the drill bit is turned into the interval, it can make the well longer. Usually between one and two miles.

Click to enlarge

(Source: Whiting)

Whiting (NYSE:WLL) provides an excellent look at well design. A lateral is completed in stages. The shorter the stages, the better the completions crew is able to create fractures in the shale. When sliding sleeves were still used (and some operators still do), there were limited entry points. The perf clusters in the plug & perf design help to break up the interval near the well bore. The greater the fractures created, the more fluids and proppant are needed. The proppant is pushed into the induced fractures, and this holds it open. A petroleum engineer could describe this better, but it provides an idea of how horizontal production works. Some of the newer frac jobs or Mega-Fracs are designed to break up the interval better. The operator also uses large volumes of proppant per foot. These bigger jobs have helped to support many companies that produce proppant, in the hopes larger jobs will keep volumes moving and keep prices up.

There are many types of proppant. This includes frac sand, ceramic coated sand and ceramic proppants. Frac sand is the cheapest, but ceramic coated sand and ceramics hold up better under stress. One well takes a very large amount of proppant. U.S. Silica (NYSE:SLCA) does a nice job of showing just how much sand has to move per well.

Click to enlarge

(Source: U.S. Silica)

The idea of 100 railcars full of sand needed for just one two-mile lateral provides a distinct image of just how much proppant will be needed to complete all of the US unconventional inventory. Lower oil prices have pushed demand down. When comparing 2014 to 2015, U.S. Silica saw an increase of 15% in the volumes of sand used per well. We think 2016 will see a larger increase.

Click to enlarge

(Source: U.S. Silica)

But total US volumes in 2015 were 90% of 2014, with rig counts down 47%. Although rig counts only provide the number of holes drilled and not the number of wells completed, it still shows the effect on the industry. It would be more fitting to track the number of completion crews.

Click to enlarge

(Source: Fairmount)

2016 could be a better year as Fairmount (NYSE:FMSA) believes proppant intensity will increase by 20% to 25%. It estimates the 2015 increase was 21%. More proppant is being used per stage, and more stages per foot. It shows an improving well design and more proppant needed for the additional fracturing.

Click to enlarge

(Source: Hi-Crush)

The Hi-Crush (NYSE:HCLP) slide above provides greater detail into how frac sand volumes have been since the downturn. 2Q15 was rough, but volumes bounced back. Hi-Crush, U.S. Silica, and Fairmount have seen volumes hold up, considering the sharp drop in prices. The only frac sand company to take a serious hit has been Emerge (NYSE:EMES).

Click to enlarge

(Source: Emerge)

Looking at Frac sand specifically, the Eagle Ford uses more than any other basin. The Eagle Ford is second per well to the Marcellus. The Eagle Ford is not as deep and well pressures lower in the shallow oil window. This allows for some wells to be all-sand fracs. The Bakken is deeper and the reason most of the historic wells have used ceramic proppant. Some Bakken operators have begun using all-sand fracs. This is the reason frac sand usage is increasing in North Dakota.

Frac sand and ceramic proppant is very different. Cost seems to be the biggest issue, although many operators find ceramics instrumental in completing deeper wells. So using frac sand as a benchmark isn't a great idea. There are many different types of sand, resin-coated sand and ceramic proppant. This is why sometimes sand and ceramics have the same use. Because costs are higher for ceramics, it is difficult to get an operator to choose it. This is one of the reasons frac sand providers are fairing better as demand has decreased. Currently, there is a fight for market share. Frac sand companies are in direct competition with ceramic proppant producers like CARBO (NYSE:CRR). We think names like U.S. Silica have a definite advantage.

Click to enlarge

(Source: Hi-Crush)

Frac sand is currently 92.3% of the proppant market. Resin coated sand is 3.6% and ceramics 4.1%. Frac sand greater demand than ceramics. Its market share has increased by 21% since 2009. Frac sand is pushing ceramic proppant out of wells throughout the United States.

Cost is the main inhibitor for the use of ceramics. Resin coated sand may be taking market share as well. Its cost makes it more attractive, plus it holds up well under higher pressures. This would depend on the play, but the slide below provides an idea of the pressures resin coated sand can tolerate.

Click to enlarge

(Source: Fairmount)

The products in light green are resin-coated sand. PowerProp can withstand closure pressures of 12,000 psi. It is almost double that of Northern White which is a popular frac sand.

As stated earlier, we are seeing better well design. This is creating more and better fractures in the source rock. The earliest designs in North Dakota show how far the industry has come. In 2008, Newfield (NYSE:NFX) completed Jorgenson 1-10H. It is a 4903-foot lateral using 5 stages. This well used 441,687 lbs. of frac sand. Roughly 400,000 lbs. were 20/40 sand and the rest mesh. One of the issues found with all sand fracs, is the induced fractures close because the sand crushes. This can mean production stops prematurely. Although this is less than a one-mile lateral and each stage was an average of 981 feet, it still has produced 174,547 BO. Jorgenson made it about 8 months before production dropped meaningfully. It is possible the fracs closed, but the well wasn't stimulated enough to begin with, and this is one of the bigger problems. That said, it still produced over 174KBO which is a good result given the lateral and stage length. I will go over some newer well designs later to provide a comparison.

It could be argued whether ceramic proppant is needed, or at least worth the cost. Newer completion techniques have produced varying results. It is also possible that less could be used and frac sand increased. The secret to a better design is reservoir contact. The resource is locked with in rock, and the more cracks or fissures made increases communication. An added bonus is seen when induced fractures come into contact with natural fractures providing contact to more square footage. In the past, operators tried to create long fracs as sleeves limited its ability to used multiple access points for fracking. Mega-fracs use large volumes of sand in many differing sizes. The smallest is pushed as deep into the frac as possible to provide conductivity through the entire channel. The bigger and more durable proppant is added until the entire fracture is propped. Ceramics are generally used last to prop open the biggest cracks. Some operators skip using ceramics at all. This includes EOG Resources (NYSE:EOG), which has most of the best performing Bakken wells of all time. Although crushing may be occurring, many of its locations are still producing well. Ceramics promote the greatest conductivity as well. The consistent shape provides larger gaps for the resource to flow through. But again, it comes down to cost.

CARBO states that 40% of all unconventional laterals drilled in the US are between 10,000 and 15,000 feet deep. Another 10% are deeper. In these wells, closure stress ranges from 7,000 to 20,000 psi. Northern White frac sand starts crushing at 6,000 psi. This argument is a good one, but our data shows many all sand wells continue to produce in a productive manner. Although better results have been seen using ceramics, there is little proof that it is economically better to use ceramics in newer Mega-fracs. Older sleeve completions may be better served using ceramics, as all sand completions did not fair as well.

Click to enlarge

(Source: CARBO)

The Eagle Ford is an interesting play as it has lower stresses in the oil dominant window. This area has a shallow depth when compared to the condensate and natural gas windows. Well pressures generally increase with depth and increased percentages of natural gas. The oil window is over 6,000 psi in some areas, but it has provided very low well costs due to all sand fracs.

It is important to understand the geology and development of the Eagle Ford to understand the significance of frac sand and ceramic usage. From a directional perspective, the Eagle Ford is deepest the south and east. The windows track from the southwest to northeast. The gas window (red) is the deepest window. All windows get deeper to the northeast. There have been mixed results throughout the play. EOG uses its large oil window leasehold as its flagship. This says a lot, as it has acreage in Northeast McKenzie County and Parshall field in the Bakken. It also has very good leasehold in the Delaware Permian. When analyzing EOG's acreage throughout the oil window, its best results have been in southwest Gonzales County. This may just be a sweet spot, as most think Karnes County is the center of the Eagle Ford core. It is important to note, EOG has all sand fracs in Gonzales County for years, and this is in the 8,000 to 10,000 psi area of CARBO's map. Even if the sand is crushing, it is not decreasing production. EOG uses huge volumes of sand. Its success may be due to breaking up the interval closer to the well bore. Most operators use 4 to 6 million lbs of proppant per 2-mile lateral. This is highly variable, and a general guideline. Below is an example of EOG's Eagle Ford Gonzales County wells.

EOG's Completion Design






32473 - 32477













































Click to enlarge

(Source: Texas RRC)

When looking at the above list it is important to focus on the proppant per foot. EOG has generally used this design with short laterals, so sand concentrations are twice that of a long lateral. If we compare to a 10,000-foot lateral using 4 million lbs, we see proppant concentrations decrease to 400 lbs/ft. This compares to the largest volumes of 2,435 lbs/ft and smallest 1,514 lbs/ft. Other operators have adopted some of these all sand heavy fracs. Since some plays like the Permian and Eagle Ford have seen success, ceramics are being pushed out. Deeper geology, or areas with higher natural gas resource may be contraindicated, but we won't know until we see its application expanded.

Click to enlarge

(Source: CARBO)

CARBO Ceramics provides the slide above. It did research comparing 12 sand and ceramic frac jobs. The sand wells saw production drop off a cliff at 6 months. The drop was substantial and not indicative of wells in this area. There could be several reasons why decline accelerated, but difficult to know since these locations are not revealed. It is possible the sand wells did not have enough or the right type of proppant. This could be the reason, as this was happening in the Bakken for wells completed in 2008 and 2009. There were reports the North Dakota wells had stopped producing, but this was due to low volumes of proppant. It is possible operators have found that large volumes of sand will crush but still maintain flow. This is especially true in the Permian.

Click to enlarge

(Source: CARBO)

The above map is an excellent representation of the Bakken. Most well development in North Dakota and Montana has been focused on the orange area. The majority of completions will be in the red area and to the east ending in the southwestern Mountrail County. Today's oil prices do not support wells west of the red position. Bakken operators must use ceramics, unless the design is a Mega-Frac.

There is a very large list of operators switching to Mega-Fracs in the Bakken. EOG Resources was the first. Enerplus (NYSE:ERF), QEP Resources (NYSE:QEP), ConocoPhillips (NYSE:COP), Halcon (NYSE:HK), Slawson, Petrohunt, and Oasis (NYSE:OAS) have moved to these completions. Volumes of proppant range from over 27 million lbs to 8 million lbs. Operator have had excellent results, and Mega-Fracs could remove the need for ceramics in parts of some plays. Of the 233 wells I reviewed, only a few wells used ceramic proppant. ConocoPhillips, Oasis, and Emerald (NYSEMKT:EOX) have all used ceramics in its Mega-Fracs. There has also been some limited use of resin coated sand.

Mega-Fracs have been very successful. EOG's Hawkeye 02-2501H has produced 613,283 BO in approximately 2 years of well life. It used 27 million pounds of sand. QEP's JOHNSON 8-5-9-4LL is a Three Forks well that has produced 217,451 BO in about one year. It used 14 million lbs of frac sand. Enerplus completed COURAGE 150-94-06A-18H using 13 million lbs of proppant. It has produced 367,427 BO in roughly a year and a half. The impressive performance of these wells could increase demand for frac sand and decrease the need for ceramic proppant. Resin coated sand could also be used less.

Sand heavy fracs are common place in the Permian. The geology seems to be more compatible and sees better initial production. Midland County may be the best acreage in the Permian. There are a significant number of operators including Pioneer, Occidental (NYSE:OXY), Exxon (NYSE:XOM), Apache (NYSE:APA), Chevron (NYSE:CVX), Encana (NYSE:ECA), RSP Permian (NYSE:RSPP), and Diamondback (NASDAQ:FANG) are just a few names with production in this county.


I listed the top 20 oil producers in Midland County. Obviously, Midland is just one county in a very large basin. I chose Midland because of its significant activity, which provides a large number of locations. Two smaller operators have core leasehold in Midland. Both have acreage in other counties of the Midland Basin.



Frac Sand lbs

Resin Sand


Lateral Feet



Spanish Trail 48-04G






Johnson Ranch






Elliott 33-01E





Parks Bell





Dusek 45-4





K-Bird 12-13





Atkins 14-11





Kathryn 41-32




Click to enlarge

The above wells were also pulled from wells completed in Upton, Reagan, and Martin counties. There are a large number of operators doing sand only fracs. Some are using resin coated sand in small volumes in concert with frac sand, but very little ceramics are being used. When operators began switching to plug & perf with large volumes of fluids and sand, Permian intervals began to out produce. Not only is initial production quite good, but declines are less than other major plays. This has been seen with all sand fracs, so CARBO's assertion that ceramic proppant is needed or production drops off may be incorrect. It is possible this occurred due to well designs with too little proppant and not as much due to type.

The last, most concerning variable refers to volumes. In 4Q15, year over year frac sand volumes decreased.

Frac Sand Producer

4Q15 Volume Decrease









Click to enlarge

The three top names with respect to volumes are around 20% lower, while EMES saw a much larger drop. This may be due more to logistics than anything else, as the others have an advantage. Some was due to EMES losing market share. CRR saw a 49% drop over the same time frame in volumes for ceramic proppant. Some of this could be decreased market share, but we believe there is a decreased need for ceramic proppant with newer well designs. The question is how hard it could hit the ceramic proppant business if correct? We expect these sand heavy frac jobs to increase going forward. If most plays can adopt this completion design, it is possible demand for ceramics could decrease significantly.

Of the 233 wells in North Dakota using 8 million lbs of proppant or more, only five used any ceramic proppant. Some of this is skewed due to EOG being the main Mega-Frac producer, but it speaks volumes when we see this many use only sand.


Total Proppant Lbs.

Frac Sand Lbs.

Ceramic Lbs.



























Click to enlarge

(Source: NDIC)

There was one well that used more than 8 million pounds of proppant that also used some resin coated sand. So there were six total that used proppant other than straight frac sand. Many operators are using different recipes to see whether production improves using small volumes of ceramics or if an all sand frac works as well. It is also interesting that once a frac used more than 9.3 million pounds of proppant, no E&P used ceramics.

Our findings provide some reasons to be bearish CARBO Ceramics. At initial view, one may think demand has dropped due to low oil prices. This is probably the case, but operators are using more proppant per well. As operators improve stimulation techniques, more proppant is needed per foot. This is why most frac sand providers have seen volumes decrease by 20% year over year. One would have expected this number to be higher, given decrease in world oil development. CARBO has seen volumes drop by 49% over the same period. Most operators are using approximately 30% more proppant per well on average. If the majority of that is going to buy sand, those providing ceramics may see less upside. Although we see oil prices improving to a high of $60/bbl this year, this will not create the pickup in activity needed for CARBO to be profitable.

More concerning is the evolution of well design. Mega-Fracs have produced very good results. These locations require much larger volumes of proppant. Initially, one would think this is advantageous to CARBO, but other changes are being seen. Many operators are beginning to switch over to all sand fracs. While much larger volumes of proppant are being used, also are the percentages of sand. These designs don't seem to need ceramics or resin coated sand. Some plays like the Permian are seeing much higher percentages of sand usage per well. If this is not specific to that basin, we could continue to see ceramic proppant volumes decrease even after completions accelerate. We believe that there is currently too much ceramic proppant capacity. This could mean CARBO Ceramics may have significant difficulty in improving its outlook.

Given the losses incurred over the last four quarters, CARBO may not have anything positive in its 12-month outlook. More importantly, demand could wane as CARBO loses market share to frac sand providers.

Disclaimer: This article is limited to the dissemination of general information pertaining to its advisory services, together with access to additional investment-related information, publications, and links. Accordingly, the publication of articles should not be construed by any consumer and/or prospective client as a solicitation to effect, or attempt to effect transactions in securities, or the rendering of personalized investment advice for compensation, over the internet. This information is provided for guidance and information purposes only. This information is not intended to provide investment, tax, or legal advice. The information contained herein has been compiled from sources deemed reliable and it is accurate to the best of our knowledge and belief. However, I cannot guarantee its accuracy, completeness, and validity and cannot be held liable for any errors or omissions. All information contained herein should be independently verified and confirmed. Hartstreet LLC does not accept any liability for any loss or damage whatsoever caused in reliance upon such information. Readers are advised that the material contained herein should be used solely for informational purposes. Information presented is for educational purposes only and does not intend to make an offer or solicitation for the sale or purchase of any specific securities, investments, or investment strategies. Investments involve risk and unless otherwise stated, are not guaranteed. Be sure to first consult with a qualified financial adviser and/or tax professional before implementing any strategy discussed herein.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.