UNITED STATES
SECURITIES AND EXCHANGE COMMISSION

FORM 10-K

x ANNUAL REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934

FOR THE FISCAL YEAR ENDED DECEMBER 31, 2003

OR

o TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934

For the transition period from ________ to ________.

COMMISSION FILE NUMBER 000-31687

EVERGREEN SOLAR, INC.

Registrant’s telephone number, including area code: 508-357-2221

Securities registered pursuant to Section 12(b) of the Act: NONE

Securities registered pursuant to Section 12(g) of the Act: COMMON STOCK, PAR VALUE $.01 PER SHARE

     Indicate by check mark whether the registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days:

     Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K is not contained herein, and will not be contained, to the best of the registrant’s knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. o

     Indicate by check mark whether the registrant is an accelerated filer (as defined in Exchange Act Rule 12b-2).

     As of March 17, 2004, there were 16,086,604 shares of the registrant’s Common Stock, $.01 par value per share, outstanding. The aggregate market value of the registrant’s voting equity held by non-affiliates as of June 30, 2003 was approximately $15 million.

DOCUMENTS INCORPORATED BY REFERENCE

The registrant intends to file a definitive proxy statement pursuant to Regulation 14A within 120 days of the end of the fiscal year ended December 31, 2003. Portions of such proxy statement are incorporated by reference into Part III of this Annual Report on Form 10-K.

Table of Contents

PART I

ITEM 1. BUSINESS:

OVERVIEW

     We develop, manufacture and market solar power products that are capable of providing reliable and environmentally clean electric power throughout the world. Our solar power products are targeted at the global solar power market. We believe our proprietary and patented solar power technologies, based on our String Ribbon^TM technology, offer significant design, cost and manufacturing advantages over competing solar power technologies. We intend to become a leading producer of high-quality solar power products by expanding our manufacturing capacity, reducing our manufacturing costs, developing innovative solar power products, increasing our distribution capabilities and pursuing strategic relationships.

     Since our formation in 1994, we have conducted research and development of advanced process and product technologies and, between 1997 and June 2001, we used pilot manufacturing facilities to refine our solar power products and manufacturing processes. Also in 1997, we began shipping small quantities of commercial products. In 2001, we expanded our manufacturing capacity capability by relocating our operations to a 56,250 square foot facility in Marlboro, Massachusetts where we constructed a new manufacturing line. We are continuing to refine, develop and commercialize a number of laboratory-demonstrated advancements in our solar power technologies, including advanced String Ribbon crystal growth, more efficient solar cells and improved solar panel designs. Since our inception, we have shipped over 50,000 solar panels for residential, commercial and industrial applications in the United States and internationally. As we continue to increase production volumes, we intend to actively expand existing and seek new distribution and marketing arrangements.

FINANCING TRANSACTION

     On May 15, 2003, the Company consummated the transaction contemplated by a Stock and Warrant Purchase Agreement which was entered into on March 21, 2003 with Perseus 2000, L.L.C., Nth Power Technologies Fund II, LP, Nth Power Technologies Fund II-A, LP, RockPort Capital Partners, L.P., RP Co-Investment Fund, I, Micro-Generation Technology Fund, LLC, UVCC Fund II, UVCC II Parallel Fund, L.P., Caisse de depot et placement du Quebec, CDP Capital — Technology Ventures U.S. Fund 2002 L.P., Beacon Power Corporation, Massachusetts Technology Park Corporation, Zero Stage Capital VII, L.P., Zero Stage Capital (Cayman) VII, L.P., Zero Stage Capital SBIC VII, L.P., IMPAX Environmental Markets plc, Merrill Lynch New Energy Technology Plc, MLIIF New Energy Fund, PNE Invest Limited, Odyssey Fund, SAM Private Equity Energy Fund LP, SAM Sustainability Private Equity LP and SAM Smart Energy. As a result of the consummation of this transaction, the Company issued 26,227,668 shares of Series A convertible preferred stock at a per share purchase price of $1.12. Additionally, Beacon Power Corporation purchased a warrant to purchase 2,400,000 shares of common stock at a per share exercise price equal to $3.37, for $100,000. A total of $29.5 million was raised as a result of the consummation of the transaction. The proceeds to the Company, net of financing expenses of approximately $849,000, were approximately $28.6 million for the transaction. The shares of Series A convertible preferred stock were initially convertible into shares of common stock on a 1-to-1 basis which is subject to adjustment to account for the payment of dividends, to take into account certain changes to the Company’s capital structure and to account for future dilutive issuances. As a condition to the obligation of the purchasers to consummate the financing transaction, the Company registered for resale the shares of common stock issuable upon the conversion of the Series A convertible preferred stock and the exercise of the warrant.

2

HISTORICAL MILESTONES

     We were incorporated in August 1994 and, to date, we have achieved the following milestones along our product development and commercialization schedule:

INDUSTRY BACKGROUND

     The electric power industry is one of the world’s largest industries. Furthermore, electricity accounts for a growing share of overall energy use. We believe that deregulation and technological innovations are creating significant opportunities for new entrants and technologies within the electric power industry, just as these changes have created similar opportunities in other regulated industries such as telecommunications, banking and transportation.

     We believe that distributed generation is one of the most promising areas for growth in the global electric power industry. Distributed generation is defined as point-of-use electricity generation that either supplements or bypasses the electric utility grid, and employs technologies such as solar power, microturbines and fuel cells. Distributed generation is expected to provide greater portability, reliability, power quality and user control. We believe capacity constraints, increased demand for power reliability and quality, and new environmental initiatives will drive the demand for distributed generation.

     We further believe that environmentally benign, locally sourced power generation will become increasingly more important for economic development, environmental policy, and national security. Increasing attention to global warming, global energy policy, and regional stability and development will support the deployment of distributed generation, particularly renewable energy.

SOLAR POWER APPLICATIONS AND BENEFITS

     Unlike many other distributed generation technologies that have been under development but not in widespread commercial use, solar power technology has had a growing worldwide market for over 25 years in the following applications:

-

On-grid. On-grid applications provide supplemental electricity to customers that are served by an electric utility grid but choose to generate a portion of their electricity needs on-site. On-grid applications have been the fastest growing part of the solar power market, largely driven by the worldwide trend toward deregulation and privatization of the electric power industry as well as by government initiatives, including incentive programs to subsidize and promote solar power systems in several countries including Japan, Germany and the United States. On-grid applications include residential and commercial rooftops and building facades and are available for both new construction and existing structures.

-

Off-grid. Off-grid applications serve markets where access to conventional electric power is not economical or physically feasible. Solar power products can provide a cost-competitive, reliable alternative for powering highway call

3

boxes, microwave stations, portable highway road signs, remote street or billboard lights, vacation homes, rural homes in developed and developing countries, water pumps and battery chargers for recreational vehicles and other consumer applications.

     Solar power has emerged as one of the primary distributed generation technologies seeking to capitalize on the opportunities resulting from trends affecting the electric power industry. Relative to other distributed generation technologies, solar power benefits include:

-

Modular and scaleable. From tiny solar cells powering a hand-held calculator to an array of roof panels powering an entire home to acres of panels on a commercial building roof or field, solar power products can be deployed in many sizes and configurations and can be installed almost anywhere in the world. Solar is among the best technologies for power generation in urban areas, environmentally sensitive areas, and geographically remote areas in both developing and developed countries.

-

Reliable. With no moving parts and no fuel supply required, solar power systems reliably power some of the world’s most sensitive applications, from space satellites to microwave stations in the mountains and other remote, harsh environments which typically require reliable power sources. Solar panels typically carry warranties as long as 25 years.

-

Dual use. Solar panels are expected to increasingly serve as both a power generator and the skin of the building. Like architectural glass, solar panels can be installed on the roofs or facades of residential and commercial buildings.

-

Environmentally cleaner. Solar power systems consume no fuel and produce no air or water emissions.

THE SOLAR POWER CHALLENGE

     Although solar power has been technically proven for over 25 years and is widely cost-effective for off-grid applications, we believe the principal challenge to widespread adoption of solar power is reducing manufacturing costs without impairing product performance or reliability. We believe the following advancements in solar power technology are necessary to meet this challenge:

-

Efficient material use. Reduce raw materials waste, particularly the waste associated with slicing silicon blocks by conventional crystalline silicon technology.

-

Simplified and continuous processing. Reduce reliance on expensive, multi-step manufacturing processes.

-

Improved product design and performance. Increase product conversion efficiency, longevity and ease of use. Conversion efficiency refers to the fraction of the sun’s energy converted to electricity.

     We further believe the two principal solar power technologies, crystalline silicon and thin films, have not adequately addressed this challenge:

     Crystalline Silicon. Crystalline silicon technology was the earliest practiced solar power technology and remains the foundation for most solar power applications. Conventional crystalline silicon technology involves slicing thin wafers from solid crystalline silicon blocks. Crystalline silicon products are known for their reliability, performance and longevity; however, factors such as high materials waste from slicing, and multi-step processing procedures have limited the ability of conventional crystalline silicon manufacturers to reduce manufacturing costs.

     Thin Films. While most major solar power manufacturers currently rely on crystalline silicon technology for the majority of their solar cell production, many are also developing alternative thin film technologies to achieve lower manufacturing costs. Thin film technology involves depositing several thin layers of silicon or more complex materials on a substrate to make a solar cell. Although thin film techniques may be able to use material more efficiently than conventional crystalline silicon, we believe the commercial acceptance of thin film technology is limited due to higher capital costs, lower manufacturing yields, lower conversion efficiency, reduced product performance and reliability, and, in some cases, concerns with toxic materials. Recent curtailments of R&D programs as well as plant closures by some industry participants, most notably BP Solar in 2002, highlight the commercial challenges that thin films have faced.

4

OUR TECHNOLOGY SOLUTION

     We believe our technologies and processes are unique among our competitors. Both our technologies and processes have been designed to reduce manufacturing costs while improving product design. We are developing technology at the wafer, cell and panel stages of manufacturing, and we hold patents and other intellectual property in all three areas. We believe our String Ribbon wafer manufacturing technology is our core technology and offers a substantial opportunity to reduce cost and otherwise advance our business through reduced materials cost, simpler processing and lower required economies of scale.

     In the String Ribbon technique, strings are pulled vertically through a shallow pool of molten silicon, and the silicon solidifies between the strings to form a continuous ribbon of crystalline silicon. The ribbon is then cut and prepared for cell fabrication. The use of strings to aid in the simplified growth of a silicon ribbon is what distinguishes our proprietary and patented String Ribbon technology from other advanced crystalline silicon wafer technologies that do not involve slicing. We believe our String Ribbon technology for the growth of solar wafers has the following significant advantages:

-

Efficient materials use. Unlike conventional bulk crystalline silicon wafer technology, in which solid blocks of silicon are sliced into thin wafers at significant expense and silicon waste, our technology grows a continuous, flat ribbon to the desired thickness. Since our technology does not involve slicing solid blocks, we can use approximately half as much silicon as conventional crystalline silicon techniques and we believe we can reduce this amount to about one-fifth in the future through production of thinner wafers.

-

Continuous processing. Our technology permits the continuous growth of crystalline silicon ribbon, which can lead to high automation, efficient equipment use and improved productivity.

-

Modularity. Dual-String Ribbon furnaces individually have a capacity of approximately 0.1 megawatts per year, whereas conventional bulk technology typically has an individual machine capability of 3 or more megawatts per year, and the trend has been to increase minimum unit size in recent years. We expect that this modularity may enable us to achieve economies of scale at a smaller scale than conventional technologies, which could enable us to manufacture internationally in more numerous, smaller factories to better serve global markets.

OUR PRODUCTS

     Solar power products in general are built-up through four stages of production:

-

Wafers. A crystalline silicon wafer is a flat piece of crystalline silicon that can be processed into a solar cell. Our rectangular wafers currently measure 81 millimeters by 150 millimeters and are approximately as thick as a business card.

-

Cells. A solar cell is a device made from a wafer that converts sunlight into electricity by means of a process known as the photovoltaic effect. Our solar cells produce approximately 1.5 watts of power each.

-

Panels. A solar panel is an assembly of solar cells that have been electrically interconnected and laminated in a physically durable and weather-tight package. A typical solar panel can produce from 20 to 200 watts of power and range in size from two to 15 square feet. A 100-watt solar panel can power a standard 100-watt light bulb, or approximately 3% of the power requirements of a typical home in the United States. Our current solar panels range from 47 to 115 watts in power.

-

Systems. A solar system is an assembly of one or more solar panels that have been physically mounted and electrically interconnected, often with batteries and/or power electronics, to produce electricity. Typical residential on-grid systems contain between 10-40 panels and produce 1-4 kW.

     We sell primarily solar panels, although we may in the future also sell wafers, cells, or systems. We believe our panels are competitive with other products in the marketplace, and some customers have commented that our panels have benefits regarding appearance, electrical design capability, and ease of use.

     Our solar panels are certified to international standards for safety and quality. If our development programs are successful, we expect to continue to increase the conversion efficiency and power of our solar panels as we expand our manufacturing capacity.

5

SALES AND MARKETING

     Market Focus

     We intend to primarily target the on-grid markets and the off-grid rural electrification market, where we believe growth prospects are the largest and where we expect our solar power technology will provide us the greatest competitive advantage. These markets are characterized as follows:

-

On-grid. The on-grid market is currently the fastest growing solar power market within which we are focused primarily on U.S. and European markets.

-

Off-grid rural electrification. Within the off-grid market, we believe that rural electrification has the largest potential but is the least penetrated market as evidenced by the two billion people in the world without conventional electricity. Marketing, financing, local infrastructure and support are projected to remain the principal challenges to greater expansion of this market. To date, we have primarily pursued sales in Central and South America and the Pacific Rim.

     Since inception, over 89% of our product sales have been to on-grid market segments, and over 65% of our product sales have been exported from the U.S. The majority of our export sales to date have been to Europe.

     Distribution, Marketing and Other Strategic Relationships

     We bring our solar power products to market using distributors, system integrators and other value-added resellers. Our resellers often add value through system design by incorporating our panels with batteries, associated electronics, structures and wiring systems. Most of our resellers have a geographic or applications focus. Our channel partners include companies that are exclusively solar distributors as well as others for whom solar power is an extension of their core business, such as, engineering design firms or other energy product marketers.

     We expect to collaborate closely with a relatively small number of resellers throughout the world. We currently have approximately 25 distributors worldwide and are actively adding new accounts and channel partners. We intend to selectively pursue additional strategic relationships with other companies worldwide for the joint marketing, distribution and manufacturing of our products. These resellers are expected to range from large, multinational corporations to small, development-stage companies, each chosen for their particular expertise. We believe that these relationships will enable us to leverage the marketing, manufacturing and distribution capabilities of other companies, explore opportunities for additional product development, and more easily and cost-effectively enter new geographic markets, attract new customers and develop advanced solar power applications.

     We currently work with a relatively small number of resellers who have particular expertise in a selected geographic or applications market segment. Sales to our ten largest resellers have accounted for approximately 82% of our total product revenues since inception. No single reseller has accounted for more than 45% of total revenues over that period. As we continue to expand manufacturing capacity and sales volumes, we anticipate developing relationships with additional resellers. During fiscal year 2003, approximately 27% of our product sales were made to resellers in the United States, and all of our research revenue was generated within the United States. During the same period, revenue from a German distributor and the National Renewable Energy Laboratory accounted for approximately 39% and 10% of total revenues, respectively.

     In addition, we market our products through trade shows, on-going customer communications, promotional material, our web site, direct mail and advertising. Our staff provides customer service and applications engineering support to our distribution partners while also gathering information on current product performance and future product requirements. Our internal sales force currently handles all solar power product sales.

MANUFACTURING

     Our principal manufacturing objective is to provide for large-scale manufacturing of our solar power products at low costs that will enable us to penetrate price-sensitive solar power markets. Our 56,250 square foot facility in Marlboro, Massachusetts includes approximately 35,000 square feet of manufacturing space. This facility includes a complete line of equipment to manufacture String Ribbon wafers, fabricate and test solar cells, and laminate and test panels. The first of the facility’s two planned manufacturing lines entered service in 2001. During 2002, we initiated the design and construction of the second production line in our Marlboro facility. At the end of 2003, equipment for parts of the cell processing and module fabrication operations began production.

6

     In addition to our current investment in our Marlboro, Massachusetts facility, we intend to selectively pursue opportunities to establish local manufacturing arrangements on a worldwide basis. Because the market opportunity for solar power encompasses numerous applications in both developed and developing nations worldwide, we expect a significant portion of our future sales will be made outside of the United States. Despite these opportunities, manufacturing of solar power products has remained largely concentrated in the United States, Europe and Japan due to factors such as reduced economies of scale and technical process complexities of establishing local manufacturing facilities.

     In spite of these barriers, we believe there are several advantages to local manufacturing, including enhanced brand recognition in local markets, avoidance of import tariffs and access to local private or public sector financing. We believe that our String Ribbon technology and our innovative manufacturing techniques offer greater advantages than other competing technologies, which we believe will enable us to establish fully integrated factories at a smaller scale that can better grow in concert with market demands. Consequently, we expect to pursue local manufacturing of our products in selected target markets. We also expect that our technologies will allow us to efficiently scale our production to take advantage of market opportunities as they arise.

RESEARCH AND DEVELOPMENT

     Because we believe continuously improving our technology is an important part of our overall strategy, we have maintained and intend to maintain a strong research and development effort. To this end, our Marlboro, Massachusetts facility has approximately 6,000 square feet dedicated to research and development and contains equipment to support the development, fabrication and evaluation of new solar power products and technologies.

     We intend to continue our policy of selectively pursuing contract research, product development and market development programs funded by various agencies of the United States, state and international governments to complement and enhance our own resources. The percentage of our total revenues derived from government-related contracts was approximately 17% for the year ended December 31, 2003. During 2003, we had two active multi-year research contracts summarized as follows (in millions):

     We have fully recognized all revenue associated with our contract with the National Institute of Standards and Technology which expired on October 31, 2003, and we expect the remaining $1.4 million of revenue associated with the contract with the National Renewable Energy Laboratory will be recognized as work is performed over the remaining life of the contract, which expires on May 31, 2005.

     These and other research contracts we have obtained generally provide for development of advanced materials and methods for wafer, cell and panel manufacturing, product development and market development. In all cases to date, we retain all rights to any intellectual property and technological developments resulting from the government funding, with the exception of government “march-in” rights to practice the technology on its own behalf and certain rights universities retain for work they perform under subcontract to us. These contracts usually require the submission by us of technical progress reports, most of which may become publicly available. These contracts are generally cost-shared between the funding agency and us with our share of the total contract cost historically ranging from approximately 30% to 70%. The contracts normally expire between six months and three years from their initiation. We recognized research revenues of $932,000 in 2001, $1.4 million in 2002, and $1.6 million in 2003 from several government-sponsored research contracts. We recorded research and development expenditures, including the cost of research revenue, of $3.1 million in 2001, $3.7 million in 2002 and $3.8 million in 2003.

7

The following table summarizes research revenues received from various government agencies as a percentage of total revenues:

INTELLECTUAL PROPERTY RIGHTS

  Patents

     We believe that our commercial success will significantly depend on our ability to protect our intellectual property rights underlying our proprietary technologies. We seek United States and international patent protection for major components of our technology platform, including our crystalline silicon wafers, solar cells and solar panels. We own or have licensed 20 issued United States patents and 3 issued foreign patents in the solar power field. These patents began to expire in 2003 and will all be expired by 2019. In addition, we have 5 United States patent applications pending. We decide whether and in what foreign countries to file counterparts of our United States patent applications. We devote substantial resources to building a strong patent position, and we intend to continue to file additional United States and foreign patent applications to seek protection for technology we deem important to our commercial success.

     Crystalline Silicon Wafers. Dr. Emanuel Sachs, a tenured Professor of Mechanical Engineering at the Massachusetts Institute of Technology, developed our core String Ribbon technology. Dr. Sachs has been awarded three issued United States patents for the String Ribbon technology. An additional issued patent for a related technology, invented by two employees of the United States National Renewable Energy Laboratory, formerly the Solar Energy Research Institute, was assigned to Dr. Sachs in 1984.

     In September 1994, Dr. Sachs granted us an irrevocable, worldwide, royalty-bearing license to practice the String Ribbon technology and related patents under a license and consulting agreement. The patents underlying this agreement began to expire in 2003 and will all have expired by the end of 2004. This agreement permits us to sublicense any of our license and other rights under the agreement. The license is exclusive worldwide, subject only to nonexclusive, nontransferable rights held by the United States Department of Energy to practice the String Ribbon technology on its own behalf. Dr. Sachs continues to actively consult with Evergreen Solar on new technological developments.

     We have been awarded 3 issued United States patents and have filed 6 patent applications on our own, internally-developed inventions related to String Ribbon and wafer fabrication, which are method inventions relating to automated, high-yield production techniques.

     Solar Cell Fabrication. We have been awarded 5 issued United States patents and 1 foreign patent relating to our solar cell processing technology. The issued United States patents relate to the method for forming wrap-around contacts on solar cells and a method for processing solar cells.

     Solar Panels. We have been awarded 8 issued United States patents and 1 foreign patent relating to advanced solar panel designs. The 8 issued United States patents relate to solar cell modules with an improved backskin, solar cell modules with an interface mounting system, an encapsulant material for solar cell modules and a solar cell roof tile system.

     The long-term status of patents in our industry, as well as others, are generally uncertain and involve complex legal and factual questions. Furthermore, even if patents are licensed or issued to us, others may design around the patented technologies. In addition, we could incur substantial costs in litigation if we are required to initiate patent litigation to enforce our patent rights, and the outcome of any patent litigation is uncertain. The following may impair our patent positions:

-

our pending patent applications may not result in issued patents;

-

the claims of patents which are issued may not provide meaningful protection;

-

we may not develop additional proprietary technologies that are patentable;

-

patents licensed or issued to us may not provide a basis for commercially viable products or may not provide us with competitive advantages and may be challenged by third parties; or

8

-

patents of others may have an adverse effect on our ability to do business.

     Although we do not believe that our technologies infringe the rights of third parties, third parties could in the future assert infringement claims against us, which may result in costly litigation or require us to obtain a license to third-party intellectual property rights. We may be unable to obtain required licenses or obtain these licenses on terms that are acceptable to us, either of which would substantially impair our business.

     Trade Secrets and Other Rights

     With respect to proprietary know-how that is not patentable and for processes for which patents are difficult to enforce, we rely on trade secret protection and confidentiality agreements to protect our interests. We believe that several elements of our solar power products and manufacturing processes involve proprietary know-how, technology or data, which are not covered by patents or patent applications including selected materials, technical processes, equipment designs, algorithms and procedures. We have taken security measures to protect proprietary know-how and technologies and confidential data, and we continue to explore additional methods of protection. While we require all employees, key consultants and other third parties to enter into confidentiality agreements with us, we cannot be assured that proprietary information will not be disclosed inappropriately, that others will not independently develop substantially equivalent proprietary information and techniques or otherwise gain access to our trade secrets, or that we can meaningfully protect our trade secrets. Any material leak of confidential information into the public domain or to third parties could result in the loss of a competitive advantage in the solar power market.

COMPETITION

     The market for solar power products is intensely competitive. There are over 20 companies in the world that produce solar power products, including BP Solar, Kyocera Corporation, Royal Dutch Shell, Sharp Corporation, RWE, and Sanyo Corporation. All of these solar power product producers, as well as several others, have historically derived all or a majority of their solar power product sales from conventional manufacturing technology that involves using wafers made from slicing solid blocks of crystalline silicon. In addition, some of these companies are developing advanced crystalline silicon or thin film technologies, including technologies such as advanced crystalline sheet and ribbon technologies and thin films of amorphous silicon, cadmium telluride and copper indium diselenide, and project future cost savings similar to or greater than ours. We believe several of our competitors are developing and are currently producing products based on crystalline silicon wafer technologies that, like String Ribbon, do not involve slicing silicon blocks.

     We believe the technologies used in our current solar power products have some similarities with those employed in our competitors’ products. However, because our solar power products are made from crystalline silicon, we believe our current products may be more reliable than those made from thin film technologies which often use materials, such as amorphous silicon, which we believe may be less stable.

     Solar power has certain advantages and disadvantages when compared to other conventional energy sources. The advantages include the ability to deploy products in many sizes and configurations, to install products almost anywhere in the world, to provide reliable power for many applications, to serve as both a power generator and the skin of a building and to eliminate air, water and noise emissions. However, unlike most conventional energy generators, which can produce power on demand, solar power cannot generate power when sunlight is not available. In addition, based on current technology, the upfront cost, including installation, of conventional energy generators is often lower than that of solar power products. However, we believe that the relative cost of power produced over the lifetime of solar products as compared with conventional energy generators often depends on the application. For example, solar power products may be more cost-effective over the long-term for remote applications that are not connected to the utility grid and that have smaller power requirements, while conventional power sources may be more cost effective for larger, grid-connected applications.

     Many of our competitors have substantially greater financial resources, research and development staff, manufacturing facilities, sales and marketing experience, distribution channels and human resources than we do. In order to compete effectively against these companies, we will need to demonstrate to potential customers and partners that our solar power products perform better, or are less expensive than those of our competitors.

     In addition, we believe that the market for solar power specifically, and electric power in general, may be subject to rapid technological development. This rapid development may result in some of our solar power products becoming obsolete before we can recover development expenses. We also expect that future competition will come not only from existing competitors, but also from new entrants to the market with new technological solutions. We may be unable to compete successfully against

9

present and future competitors and our failure to successfully compete could significantly reduce our market share, our revenues and our prospects for profitability.

ENVIRONMENTAL REGULATIONS

     We use, generate and discharge toxic, volatile or otherwise hazardous chemicals and wastes in our research and development and manufacturing activities. We are subject to a variety of federal, state and local governmental regulations related to the storage, use and disposal of hazardous materials.

     We believe that we have all environmental permits necessary to conduct our business. We believe that we have properly handled our hazardous materials and wastes and have not contributed to any contamination at any of our past or current premises. We are not aware of any environmental investigation, proceeding or action by federal or state agencies involving our past or current facilities. If we fail to comply with present or future environmental regulations, we could be subject to fines, suspension of production or a cessation of operations. Any failure by us to control the use of or to restrict adequately the discharge of hazardous substances could subject us to substantial financial liabilities, operational interruptions and adverse publicity, any of which could materially and adversely affect our business, results of operations and financial condition. In addition, under some federal and state statutes and regulations, a governmental agency may seek recovery and response costs from operators of property where releases of hazardous substances have occurred or are ongoing, even if the operator was not responsible for the release or otherwise was not at fault.

EMPLOYEES

     As of December 31, 2003, we had 151 full-time employees, including 18 engaged in research and development and 118 engaged in manufacturing. Nineteen of our employees have advanced degrees, including 5 with Ph.D.s. None of our employees are represented by any labor union nor are they organized under a collective bargaining agreement. We have never experienced a work stoppage and believe that our relations with our employees are good.

AVAILABLE INFORMATION

     Our annual report on Form 10-K, quarterly reports, current reports on Form 8-K, and all amendments to those reports are made available free of charge though our internet website (http://www.evergreensolar.com) as soon as practicable after such material is electronically filed with, or furnished to, the Securities and Exchange Commission.

ITEM 2. PROPERTIES:

     Our headquarters is currently located in a leased space in Marlboro, Massachusetts, where we currently occupy approximately 56,250 square feet of administrative, laboratory and manufacturing space. Our lease expires on June 30, 2010.

     On January 24, 2004, we signed a new lease for 23,839 square feet of additional warehouse and office space located in Marlboro, Massachusetts. This lease expires on January 24, 2010.

ITEM 3. LEGAL PROCEEDINGS:

     We are not a party to any material legal proceedings.

ITEM 4. SUBMISSION OF MATTERS TO A VOTE OF SECURITY HOLDERS:

     No matters were submitted to a vote of security holders during the quarter ended December 31, 2003.

10

PART II

ITEM 5. MARKET FOR THE COMPANY’S COMMON EQUITY AND RELATED STOCKHOLDER MATTERS:

Market for Our Common Stock

     Our common stock is traded on the Nasdaq National Market under the symbol “ESLR”. The following table sets forth for the calendar periods indicated, the high and low sales price of our common stock on the Nasdaq National Market.

     On March 17, 2004 there were approximately 176 stockholders of record of our common stock. We have never declared or paid cash dividends on shares of our common stock. Holders of shares of Series A convertible preferred stock are entitled to a quarterly dividend at an annual rate of 10% payable in cash or, at our option, we may accrete the Series A convertible preferred stock 10% dividend to the conversion price and liquidation preference of the Series A convertible preferred stock. Apart from the dividend on the Series A convertible preferred stock, we intend to retain any earnings for use in our business.

     Please see Part III, Item 12: Securities Ownership of Certain Beneficial Owners and Management for the information regarding securities authorized for issuance under our equity compensation plans.

11

ITEM 6. SELECTED FINANCIAL DATA:

     You should read the data set forth below in conjunction with our financial statements and related notes and “Management’s Discussion and Analysis of Financial Condition and Results of Operations” appearing elsewhere in this filing. The statement of operations data presented below for the fiscal years ended December 31, 2001, 2002, and 2003 and the balance sheet data at December 31, 2002 and 2003 have been derived from our audited financial statements which appear elsewhere in this filing. The statement of operations data presented below for the years ended December 31, 1999 and 2000, and the balance sheet data at December 31, 1999, 2000 and 2001 have been derived from our audited financial statements, which are not included in this filing.

12

ITEM 7. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS:

     We caution readers that statements in this Annual Report on Form 10-K that are not strictly historical statements, including, but not limited to: statements reflecting our expectations regarding the timing, cost, and success of our manufacturing scale-up at our new facility in Marlboro, Massachusetts and future manufacturing expansion and production, as well as related financing requirements; future financial performance; our technology and product development, cost and performance; our current and future strategic relationships and future market opportunities; and our other business and technology strategies and objectives, constitute forward-looking statements which are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. These statements may be identified with such words as “we expect”, “we believe”, “we anticipate” or similar indications of future expectations. These statements are neither promises nor guarantees and involve risks and uncertainties, which could cause our actual results to differ materially from such forward-looking statements. Such risks and uncertainties may include, among other things, those risks and uncertainties described in this Annual Report and in our other filings with the Securities and Exchange Commission, copies of which may be accessed through the SEC’s Web Site at http://www.sec.gov. We caution readers not to place undue reliance on any forward-looking statements contained in this Annual Report, which speak only as of the date of this Annual Report. We disclaim any obligation to publicly update or revise any such statements to reflect any change in our expectations, or events, conditions, or circumstances on which any such statements may be based, or that may affect the likelihood that actual results will differ from those set forth in such forward-looking statements.

EXECUTIVE OVERVIEW

     We develop, manufacture and market solar power products for the global marketplace. Solar cells are semiconductor devices that convert sunlight into electricity and form the building block for all solar power products. To date, our product sales have been primarily solar panels, which have been used to generate electricity for on-grid and off-grid applications. Off-grid applications have included the electrification of rural homes, lighting for small, rural schools and power supplies for water pumping. More recently, the substantial majority of our products have been used by on-grid customers as a clean, renewable source of alternative or supplemental electricity.

During 2003, our product sales were constrained by our manufacturing capacity. During the second half of 2003, we made significant progress on our capital expansion program aimed at roughly quadrupling our production capacity by adding a second manufacturing line, which we expect to be completed by the end of 2004. We completed a $29.5 million private equity financing on May 15, 2003 which will help fund this expansion. Since product revenue has been capacity limited, the planned expansion should enable significant revenue growth during 2004 and should also further improve our product gross margins.

In anticipation of the increase in product availability, we strengthened our sales and marketing organization with three hires during the fourth quarter of 2003, and are focused on expanding sales channels. The combination of recent governmental initiatives in Germany aimed at accelerating the adoption of solar power for residential and commercial applications, our proactive sales efforts, and our capacity expansion efforts should provide us with a foundation for product revenue growth in 2004.

     Implementation of our dual-ribbon growth technology, increased production volume, adding key positions to staff across the organization, expanding and establishing key customer relationships, and consummating the Series A convertible preferred stock and warrant financing were the key factors in determining our performance for 2003. We demonstrated the dual-ribbon growth technology (pulling two string ribbons out of one furnace) on a limited production basis during the fourth quarter in 2003, which we expect to substantially lower the manufacturing cost of growing silicon wafers and roughly double the capacity of one string ribbon furnace. We expect the new crystal growth furnaces we will be installing throughout 2004 will all be dual-ribbon growth furnaces. We added several key staff in sales, administration, manufacturing and research and development during 2003 to better position ourselves for the anticipated growth of the company over the next 12 months. Most notably, we hired Richard M. Feldt as our new President and CEO in December 2003. Additionally, our marketing and sales organization was expanded during the fourth quarter to develop our marketing and distribution strategy in anticipation of future growth.

The cash raised as a result of the Series A convertible preferred stock and warrant financing provided us with the resources necessary to increase our capital spending program. However, our current cash, cash equivalents and short-term investments will not be sufficient to fund this capacity expansion and our expected increased level of operations through fiscal year 2004 and, as a result, we will need to raise significant additional financing or secure a working capital line of credit in order to successfully fund such expanded operations. If we are not able to complete a financing or secure a working capital line of credit in a timely manner, we will need to implement fundamental changes to our business and operations which will likely include substantially reducing, suspending, or terminating our capacity expansion and substantially reducing our daily operating expenditures from current levels, in which case we believe our cash, cash equivalents and short-term investments will then be sufficient to fund our operations through the end of fiscal year 2004.

13

We may need additional financing to execute our business plan sooner if we need to respond to business contingencies such as the need to enhance our operating infrastructure, respond to competitive pressures and acquire complementary businesses or necessary technologies. We do not know whether we will be able to raise additional financing or financing on terms favorable to us. If adequate funds are not available or are not available on acceptable terms, our ability to fund our operations, develop and expand our manufacturing operations and distribution network, or otherwise respond to competitive pressures would be significantly limited.

CRITICAL ACCOUNTING POLICIES AND ESTIMATES

     The preparation of consolidated financial statements in accordance with generally accepted accounting principals requires us to make estimates and judgments that affect the reported amounts of assets, liabilities, revenues and expenses, and related disclosure of contingent assets and liabilities. We base our estimates on historical experience and on various other assumptions that are believed to be reasonable under the circumstances, the results of which form the basis for making judgments about the carrying values of assets and liabilities that are not readily apparent from other sources. Actual results may differ from these estimates under different assumptions or conditions. We believe the following critical accounting policies affect our more significant judgments and estimates used in the preparation of our consolidated financial statements:

Revenue Recognition and Allowance for Doubtful Accounts

     We recognize product revenue if persuasive evidence of an arrangement exists, shipment has occurred, risk of loss has transferred to the customer, sales price is fixed or determinable, and collectibility is reasonably assured. The market for solar power products is emerging and rapidly evolving. We currently sell our solar power products primarily to distributors, system integrators and other value-added resellers within and outside of North America, which typically resell our products to end users throughout the world. For new customers requesting credit, we evaluate creditworthiness based on credit applications, feedback from provided references, and credit reports from independent agencies. For existing customers, we evaluate creditworthiness based on payment history and known changes in their financial condition.

     We also evaluate the facts and circumstances related to each sales transaction and consider whether risk of loss has not passed to the customer upon shipment. We consider whether our customer is purchasing our product for stock, and whether contractual or implied rights to return the product exist or whether our customer has an end user contractually committed. To date, we have not offered rights to return our products other than for normal warranty conditions.

     We maintain allowances for doubtful accounts for estimated losses resulting from the inability of our customers to make required payments. If the financial condition of our customers were to deteriorate, such that their ability to make payments was impaired, additional allowances could be required.

     Revenue from research grants is recognized as services are rendered to the extent of allowable costs incurred.

Inventory

     Inventory is valued at the lower of cost or market. Certain factors may impact the realizable value of our inventory including, but not limited to, technological changes, market demand, changes in product mix strategy, new product introductions and significant changes to our cost structure. Given our current production levels and the market value of our products, we currently sell our finished goods inventory at prices that are below the sum of our fixed and variable costs per unit. Accordingly, we write down our finished goods inventory to realizable value equal to the difference between the cost of inventory and the estimated market value. In addition, estimates of reserves are made for obsolescence based on the current product mix on hand and its expected net realizability. If actual market conditions are less favorable or other factors arise that are significantly different than those anticipated by management, additional inventory write-downs or increases in obsolescence reserves may be required. We treat lower of cost or market adjustments and inventory reserves as an adjustment to the cost basis of the underlying inventory. Accordingly, favorable changes in market conditions are not recorded to inventory in subsequent periods.

Warranty

     We provide for the estimated cost of product warranties at the time revenue is recognized. Given our limited operating history, we use historical industry solar panel failure rates as the basis for our warranty provision calculation. While we engage in product quality programs and processes, including monitoring and evaluating the quality of our component suppliers, our warranty obligation is affected by product failure rates and material usage and service delivery costs incurred in correcting a

14

product failure. If our actual product failure rates, material usage or service delivery costs differ from our estimates, revisions to the estimated warranty liability would be required. Since we have a limited operating history and our manufacturing process differs from industry standards, our experience may be different from the industry data used as a basis for our estimate. While our methodology takes into account these uncertainties, adjustments in future periods may be required as our products mature.

Long-lived Assets

     Our policy regarding long-lived assets is to evaluate the recoverability or usefulness of these assets when the facts and circumstances suggest that these assets may be impaired. This analysis relies on a number of factors, including changes in strategic direction, business plans, regulatory developments, economic and budget projections, technological improvements, and operating results. The test of recoverability or usefulness is a comparison of the asset value to the undiscounted cash flow of its expected cumulative net operating cash flow over the asset’s remaining useful life. If such a test indicates that an impairment is required, then the asset is written down to its estimated fair value. Any write-downs would be treated as permanent reductions in the carrying amounts of the assets and an operating loss would be recognized. To date, we have had recurring operating losses and the recoverability of our long-lived assets is contingent upon executing our business plan that includes further reducing our manufacturing costs and significantly increasing sales. If we are unable to execute our business plan, we may be required to write down the value of our long-lived assets in future periods.

Income Taxes

     We are required to estimate our income taxes in each of the jurisdictions in which we operate as part of our consolidated financial statements. This involves estimating the actual current tax in addition to assessing temporary differences resulting from differing treatments for tax and financial accounting purposes. These differences together with net operating loss carryforwards and tax credits may be recorded as deferred tax assets or liabilities on the balance sheet. A judgment must then be made of the likelihood that any deferred tax assets will be recovered from future taxable income. To the extent that we determine that it is more likely than not that deferred tax assets will not be utilized, a valuation allowance is established. Taxable income in future periods significantly different from that projected may cause adjustments to the valuation allowance that could materially increase or decrease future income tax expense.

Results of Operations

     Product revenues. Product revenues consist of revenues from the sale of solar cells, panels and systems. Product revenues represented 83% of total revenues for the year ended December 31, 2003, 79% of total revenues for the year ended December 31, 2002 and 62% of total revenues for the year ended December 31, 2001. International product sales accounted for approximately 73%, 68% and 35% of total product revenues for the years ended December 31, 2003, 2002 and 2001, respectively. During 2003, product sales to a European distributor accounted for approximately 47% of product revenue, and sales to another European distributor accounted for approximately 10% of product revenue. During 2002, sales to a European distributor accounted for approximately 56% of product revenue. During 2001, sales to a U.S. distributor and a Japanese distributor accounted for approximately 52% and 22% of product revenue, respectively. We anticipate that international sales will continue to account for a significant portion of our product revenues for the foreseeable future. Through December 31, 2003, all product revenues were denominated in United States dollars. Foreign exchange rate fluctuations have impacted the relative competitiveness of our products in international markets, but we have not had any direct foreign exchange exposure. However, we expect that during 2004, market conditions will require us to denominate a portion of our sales in local currencies, which could expose us directly to foreign exchange gains or losses. As we expand our manufacturing operations and distribution network internationally, our exposure to fluctuations in currency exchange rates may increase.

     Research revenues. Research revenues consist of revenues from various state and federal government agencies to fund our ongoing research, development, testing and enhancement of our products and manufacturing technology. We have not in the past, nor is it our intention in the future, to pursue contracts that are not part of our ongoing research activities. We recognize research revenues as services are rendered. During 2003, we had two active multi-year research contracts summarized as follows (in millions):

15