DOE EV Charging Station Data: The Federal Database Behind 180,000 US Alternative Fuel Stations

The Department of Energy Alternative Fuels Station Locator database tracks every publicly accessible electric vehicle charging station, hydrogen station, propane station, CNG station, and other alternative fuel outlet in the United States — 180,000+ stations as of 2024, with real-time status for DCFC fast chargers, providing the most comprehensive federal dataset on EV charging infrastructure deployment.

What the AFDC Is and Who Runs It

The Alternative Fuels Station Locator is one component of the Alternative Fuels and Advanced Vehicles Data Center, universally known as the AFDC, operated by the National Renewable Energy Laboratory under contract to the Department of Energy. NREL's Transportation and Mobility Research group maintains the database at afdc.energy.gov, which serves as the public-facing portal for both the station locator and a broader library of alternative fuels policy, fleet guidance, and technical resources.

The AFDC was established in 1991, when the primary policy concern around alternative fuels was natural gas vehicles and the Clean Air Act Amendments of 1990. At its founding, the database tracked compressed natural gas and liquefied natural gas stations, propane outlets, and the early generation of ethanol and methanol fueling sites. Electric vehicle coverage was modest through the 2000s — the number of publicly accessible charging stations in the United States was in the hundreds — but grew rapidly after 2010 as the Nissan Leaf and Chevrolet Volt commercialized mass-market EVs and the DOE's EV Project deployed thousands of Level 2 chargers in pilot markets.

NREL maintains the AFDC in coordination with the Clean Cities Coalition Network, a program of more than 100 regional coalitions across the country that work with fleets, fuel providers, governments, and communities to accelerate alternative fuel adoption. Clean Cities coalitions serve as local data contributors and validators, reporting new station openings, closures, and corrections in their regions. Major charging networks — ChargePoint, EVgo, Blink, Electrify America, and others — submit data directly via API integrations, and NREL's staff conduct ongoing data quality reviews. The AFDC API is available free of charge after registration at developer.nrel.gov, where it sits alongside NREL's solar irradiance, wind resource, and energy incentives APIs.

Station Types and Counts

The AFDC tracks seven alternative fuel types, each with a distinct infrastructure footprint and trajectory:

• Electric (ELEC) — The dominant and fastest-growing category. As of 2024, approximately 64,000 public charging stations exist in the United States, but the AFDC reports at the port/connector level as well, reaching more than 180,000 individual charging outlets. The distinction matters: a single station may have dozens of connectors; DCFC fast-charging stations from Electrify America routinely have 12 to 30 individual ports per location. Electric stations are further segmented by charge level (Level 1, Level 2, and DC Fast Charge) and network affiliation.
• Hydrogen (HY) — Approximately 60 stations, heavily concentrated in California where the California Energy Commission has funded most of the national hydrogen fueling network. Hydrogen stations serve fuel cell electric vehicles (FCEVs) such as the Toyota Mirai and Hyundai Nexo. Station reliability has been a persistent challenge; extended outages at California hydrogen stations have affected FCEV owners severely, since there is no home fueling option for hydrogen.
• Compressed Natural Gas (CNG) — Approximately 900 public stations, supplemented by a much larger base of private or limited-access fleet stations. CNG fuels heavy-duty transit buses, refuse trucks, and long-haul trucking applications where natural gas offers a cost advantage over diesel. Deployment has been stable rather than growing, as fleet operators weigh CNG infrastructure investment against the advancing economics of battery-electric for medium-duty and hydrogen fuel cell for heavy-duty applications.
• Liquefied Natural Gas (LNG) — Approximately 100 stations, almost entirely oriented toward over-the-road trucking where the higher energy density of liquefied versus compressed gas extends range. LNG stations require cryogenic storage equipment and specialized dispensing hardware. The sector has seen limited growth as battery-electric trucks have captured increasing industry attention.
• Propane (LPG) — Approximately 2,800 stations, primarily serving forklifts, farm equipment, school buses, and light-duty fleet vehicles in rural and agricultural regions. Propane dispensing infrastructure at existing rural fueling stations is relatively low-cost to add, giving propane better rural coverage than electric in some geographies.
• Biodiesel (BD) — Approximately 1,700 stations offering B20 or higher biodiesel blends. Biodiesel is drop-in compatible with diesel engines and does not require vehicle modifications; the AFDC tracks retail fuel sites offering biodiesel blends rather than dedicated alternative fuel infrastructure.
• E85 Ethanol (E85) — Approximately 4,300 stations, primarily in the Midwest corn belt. E85 requires flex-fuel vehicles capable of operating on up to 85 percent ethanol blended with gasoline. The E85 station network is the largest non-electric alternative fuel network by station count and has been relatively stable in recent years as the flex-fuel vehicle market has matured.

Data Attributes

Each AFDC station record contains approximately 80 fields covering location, access conditions, fuel-specific characteristics, and operational metadata. The core fields available through the API and flat-file downloads include:

The combination of ev_level1_evse_num, ev_level2_evse_num, and ev_dc_fast_num allows analysis at the port level rather than the station level, which matters substantially for infrastructure capacity calculations. A single station record with ev_dc_fast_num of 28 represents 28 simultaneous fast charging sessions — closer to a highway travel plaza than to the single-port Level 2 charger at a hotel parking space.

EV Charging Networks

The AFDC ev_network field maps each station to its operating network, which matters because network affiliation determines payment method, app compatibility, pricing structure, and real-time availability visibility. The major networks as tracked by the AFDC include:

• Tesla / Tesla Supercharger — Tesla's proprietary DC fast charging network, historically restricted to Tesla vehicles but opened to other EVs beginning in 2022 in Europe and 2023 in the United States. Tesla Supercharger locations began appearing in the AFDC after Tesla opened its network, making previously uncounted infrastructure visible in the federal dataset for the first time. Supercharger stalls typically deliver 150–250 kW per stall at V3 sites.
• ChargePoint — The largest US charging network by station count, primarily Level 2 commercial stations at workplaces, retail, and municipal locations. ChargePoint operates a network model where site hosts own the hardware and ChargePoint provides the software platform and network connectivity. ChargePoint data feeds directly into the AFDC and represents the largest contributor of Level 2 station records.
• Electrify America — A DC fast charging network established by Volkswagen as part of the settlement from the 2015 diesel emissions fraud. Electrify America is required to invest $2 billion in EV infrastructure through 2026; its network is concentrated along major interstate corridors with high-power stations (up to 350 kW per stall) capable of delivering the fastest charging speeds available to compatible vehicles.
• EVgo — A publicly traded DC fast charging network operator focused on urban and suburban markets, often co-located with grocery and retail chains. EVgo has partnered with General Motors on deployment targets and operates a significant share of the urban DCFC network outside of Tesla.
• Blink — A mix of Level 2 and DC fast chargers, including a large legacy installed base of Level 2 stations at hotels and commercial properties dating to the early 2010s. Blink has had a more complex operational history than ChargePoint or EVgo, with periods of ownership transition that affected station uptime.
• BP Pulse (formerly Greenlots, then Shell Recharge, now BP Pulse in the US market following BP's acquisitions) and EV Connect represent additional network operators with meaningful regional footprints that appear in the AFDC data.

Non-networked stations — Level 2 chargers installed at properties without network management software — appear in the AFDC as “Non-Networked” and represent a significant share of the Level 2 installed base, particularly at older installations and municipal facilities. Non-networked stations provide no real-time availability data and may have lower uptime visibility.

The NEVI Formula Program

The National Electric Vehicle Infrastructure Formula Program, universally called NEVI, is the largest dedicated federal investment in EV charging infrastructure in US history. Enacted as part of the Infrastructure Investment and Jobs Act of 2021 (IIJA, also known as the Bipartisan Infrastructure Law), NEVI directs $5 billion over five fiscal years (2022 through 2026) to states for deployment of EV charging infrastructure along designated Alternative Fuel Corridors.

NEVI is administered by the Federal Highway Administration, which sets minimum technical standards that all NEVI-funded stations must meet. The core NEVI standards as established by FHWA are:

• Corridor spacing — NEVI-funded stations must be located within one mile of an Interstate highway exit and spaced no more than 50 miles apart along the designated Alternative Fuel Corridor. The 50-mile spacing requirement is designed to eliminate range anxiety for typical battery electric vehicles by ensuring that drivers on Interstate routes always have a charging stop within half a charge's drive.
• Minimum charger count and power — Each NEVI station must have at least four DC fast charger ports with a minimum output of 150 kW per port. The 150 kW threshold was chosen to accommodate a broad range of current EVs; most modern battery electric vehicles accept 150 kW or less at peak, while higher-capacity vehicles can benefit from the full throughput.
• Uptime requirement — NEVI stations must maintain a minimum 97 percent uptime, measured as the percentage of time each individual charger port is operational and available. The 97 percent standard was substantially more stringent than the typical uptime performance of the US charging network at the time the rule was issued, and it required network operators to invest in remote monitoring, predictive maintenance, and on-site service response capabilities.
• Connector standard — Initially, NEVI required both CCS and CHAdeMO connectors. FHWA updated its NEVI standards in 2024 to require NACS (the North American Charging Standard, now codified as SAE J3400) following the industry's rapid adoption of NACS across major automakers. New NEVI-funded stations are required to include NACS connectors; existing stations must add NACS capability within a transition period.
• Payment and accessibility — NEVI stations must accept payment by credit or debit card (without requiring a network membership) and must comply with ADA accessibility requirements for the charger hardware, parking surfaces, and approach paths.

NEVI funds are distributed to states by formula based on interstate lane miles and are implemented through state DOT plans reviewed and approved by FHWA. States were required to submit EV Infrastructure Deployment Plans outlining their proposed corridor-by-corridor deployment priorities, station siting criteria, and procurement approaches. As NEVI stations are deployed, their records should appear in the AFDC dataset tagged with network information and meeting the minimum technical attributes specified in the FHWA standards. The AFDC is the federal tracking mechanism for understanding NEVI deployment progress at the station level.

Connector Standards: J1772, CHAdeMO, CCS, and NACS

The connector standard landscape for EV charging in the United States has undergone a significant consolidation since 2020, moving from a three-way split among J1772, CHAdeMO, and CCS toward a two-standard world of CCS and NACS, with J1772 remaining universal for AC Level 2 charging.

SAE J1772 (the “J plug”) is the universal AC charging standard for Level 1 and Level 2 charging in North America. Every battery electric vehicle sold in the United States accepts J1772 for AC charging; even Tesla vehicles shipped with a J1772-to-NACS adapter from the factory. J1772 handles up to 19.2 kW AC at Level 2, sufficient for overnight home charging or workplace dwell-time charging. J1772 is not used for DC fast charging — it is exclusively an AC connector.

CHAdeMO is a DC fast charging standard developed by a consortium of Japanese automakers (Toyota, Nissan, Mitsubishi, and Honda) and first deployed commercially around 2010 alongside the Nissan Leaf. CHAdeMO was the dominant DC fast charging connector in the US market from roughly 2012 to 2016. Its decline began as US and European automakers standardized on CCS, and it has accelerated rapidly since 2020. Nissan, the primary US CHAdeMO vehicle seller, announced that the Leaf would not continue in the US market and that future Nissan EVs would use NACS. CHAdeMO connectors are being removed from newer stations or not installed in new deployments; CHAdeMO counts in the AFDC are declining as the installed base ages out.

CCS (Combined Charging System), formally SAE J1772 Combo or SAE Combo, adds two DC pins below the standard J1772 AC pins to create a combined AC/DC connector. CCS became the mandatory standard for DC fast charging in non-Tesla EVs sold in the United States from roughly 2014 onward, endorsed by the SAE, the major US automakers, and most European automakers. CCS supports charging up to 350 kW in its Combo 1 (North American) variant. For several years, CCS was the clear heir to CHAdeMO as the dominant US DC fast standard.

NACS (North American Charging Standard) is Tesla's connector standard, internally known as the “Magic Dock” hardware and externally as the Tesla connector. In November 2022, Tesla announced it was opening the NACS specification for adoption by other manufacturers. In 2023, Ford and General Motors announced they would adopt NACS for their future EV lineups, followed rapidly by virtually every major automaker selling in the US market. The SAE codified NACS as SAE J3400 in 2023. FHWA incorporated NACS/J3400 into NEVI requirements in 2024. The practical consequence for the AFDC dataset is that stations previously recorded with only CCS connectors are adding NACS hardware, and new stations are being deployed with NACS as the primary fast charging standard alongside CCS for legacy compatibility during the transition period.

The ev_connector_types field in the AFDC records this evolution in real time. A station that opened in 2016 may show only CHADEMO and J1772COMBO (the AFDC code for CCS); a station opening in 2025 is likely to show NACS and J1772COMBO, with CHAdeMO absent. Tracking the connector mix through the AFDC dataset provides a direct measure of the NACS transition rate.

Coverage Gaps and Equity Concerns

The AFDC dataset, when mapped, reveals sharp geographic disparities in EV charging access that mirror broader patterns of infrastructure investment and economic inequality. The concentration of public EV charging in urban and suburban areas — particularly on the coasts and in prosperous metropolitan cores — reflects both where EV ownership is highest and where commercial charging deployment economics work.

Rural counties, particularly those in the Great Plains, the Mountain West, and the deep South, have minimal public charging coverage outside of Interstate corridors. A driver in a rural Texas county who does not own a home with a garage for Level 1 or Level 2 home charging may have no practical access to public EV charging within a reasonable distance. This is not a theoretical concern: Census data shows that apartment and multi-family housing residents — who cannot install home chargers — are disproportionately lower-income and disproportionately located in dense urban neighborhoods where street parking predominates over garage parking.

Tribal lands present a particularly acute coverage gap. AFDC data shows that tribal reservation areas, many of which lack robust electricity grid infrastructure, have essentially no public EV charging. The federal government has recognized this through the DOE's Justice40 initiative, which directs that 40 percent of the benefits of certain federal clean energy investments flow to disadvantaged communities. NEVI program guidance incorporates Justice40 principles, and FHWA has encouraged state DOTs to extend NEVI-adjacent deployments to tribal lands, but the baseline coverage remains thin.

Research using AFDC data linked to census demographic data consistently finds that EV charger access correlates with household income at the zip code level. Higher-income zip codes have substantially more chargers per capita than lower-income zip codes controlling for EV ownership rates. The policy debate around this finding divides between the view that charger deployment follows demand (and demand follows EV ownership, which correlates with income) and the view that charger scarcity in low-income areas is itself a barrier to EV adoption that creates a self-reinforcing cycle.

Workplace charging represents a distinct gap category. Employer-provided Level 2 charging at workplaces is common at technology companies and large corporations with sustainability commitments, but is rare at blue-collar workplaces, warehouses, distribution centers, and service-sector employers. AFDC data on workplace charging is incomplete because many employer installations are classified as “private” and may not be consistently reported. The actual workplace charging population is likely larger than the AFDC record count suggests, but it is inaccessible to non-employees and thus does not address the equity gap.

Data Freshness and Real-Time Status

The AFDC operates on multiple update timescales depending on the station type and network affiliation. For networked DC fast chargers, real-time availability status (available, occupied, or out of service per port) is accessible through the AFDC API for stations whose network operators provide live data feeds. This real-time layer represents a significant capability — a developer querying for available DCFC stations near a highway exit can receive current port-level availability, not just the static station record.

Tesla Supercharger data began flowing into the AFDC after Tesla opened its network to non-Tesla vehicles in 2023. The AFDC previously had no visibility into the Supercharger network, which was the largest and most reliable DCFC network in the country by most measures. The inclusion of Supercharger data substantially improved AFDC's completeness as a fast charging coverage map, particularly along Interstate corridors where Superchargers were deployed well ahead of the Electrify America and EVgo networks.

ChargePoint, EVgo, and Blink provide data feeds to NREL that are incorporated into AFDC updates on a regular basis. Station openings, closures, and hardware changes are reflected in the dataset typically within days to weeks, depending on the network operator's reporting cadence. For non-networked stations and stations submitted by Clean Cities coalitions or individual reporters, updates are less frequent and depend on manual submission and review.

Known data quality issues in the AFDC include closed stations that remain in the dataset as “open” status after operators fail to report closures, incorrect or outdated operating hours in the access_days_time field (which is a freeform text field not validated against actual network status), and inconsistent connector type reporting for stations that have undergone hardware upgrades. The updated_at timestamp on each record indicates when the station was last modified in the AFDC, which can serve as a proxy for record freshness, though it does not guarantee that all fields reflect current conditions.

Python: Querying the AFDC API for EV Station Data

The following script uses the NREL AFDC API to retrieve EV station records for a single state, compute network and charger type distributions, and summarize access type breakdowns. The DEMO_KEY API key is available for low-volume testing without registration; production use should register for a free key at developer.nrel.gov to avoid rate limiting. The API returns a JSON envelope with total_results for pagination and alt_fuel_stations as the records array.

import requests, pandas as pd

# DOE AFDC API — register free at developer.nrel.gov
# API docs: https://developer.nrel.gov/docs/transportation/alt-fuel-stations-v1/
API_KEY = "DEMO_KEY"  # replace with your key (free registration)
base = "https://developer.nrel.gov/api/alt-fuel-stations/v1.json"

# Get all EV stations in Texas
params = {
    "api_key": API_KEY,
    "fuel_type": "ELEC",
    "state": "TX",
    "status": "E",      # E = open/available
    "limit": 200,
}
resp = requests.get(base, params=params, timeout=20)
data = resp.json()
stations = data.get("alt_fuel_stations", [])
print(f"Texas EV stations: {data.get('total_results', 0)} total")

df = pd.DataFrame(stations)
if not df.empty:
    # Network distribution
    net_counts = df["ev_network"].value_counts().head(10)
    print("\nTop EV networks in Texas:")
    print(net_counts.to_string())

    # Fast charger counts
    df["ev_dc_fast_num"] = pd.to_numeric(df.get("ev_dc_fast_num", 0), errors="coerce").fillna(0)
    df["ev_level2_evse_num"] = pd.to_numeric(df.get("ev_level2_evse_num", 0), errors="coerce").fillna(0)
    print(f"\nTotal DC fast charger ports (sample): {df['ev_dc_fast_num'].sum():.0f}")
    print(f"Total Level 2 EVSE ports (sample): {df['ev_level2_evse_num'].sum():.0f}")

    # Access type
    access_dist = df["access_code"].value_counts()
    print("\nAccess type distribution:")
    print(access_dist.to_string())

The limit parameter caps results per request; the full Texas EV station inventory exceeds 10,000 stations as of 2024, so production analysis should paginate using offset and limit until the total record count is exhausted. To retrieve all fuel types in a single request, omit the fuel_type parameter. The status filter accepts E for open stations, P for planned, and T for temporarily unavailable; omitting it returns all status codes. To filter by proximity rather than state, replace state with latitude, longitude, and radius parameters. The ev_network, ev_connector_types, and ev_charging_level parameters allow pre-filtered queries for specific network operators, connector types, or charging levels without post-processing in pandas.

EV charging deployment is one component of the broader federal energy transition infrastructure data ecosystem. For renewable energy resource mapping and generation capacity data that feeds both EV grid impact modeling and siting analysis, see USGS Wind and Solar Energy: The Federal Dataset Behind Renewable Resource Mapping.

The Energy Information Administration tracks retail electricity consumption and price data by state and sector, providing the supply-side context for understanding how EV charging load growth interacts with grid capacity and wholesale power prices. See EIA Energy Data: The Federal Database Behind Oil Prices, Natural Gas Storage, and Electricity Generation.

NEVI program station requirements are enforced through Federal Highway Administration standards; FHWA also maintains the National Bridge Inventory, the federal database covering the condition and inspection history of every highway bridge in the United States. See FHWA National Bridge Inventory: The Federal Database Behind 620,000 US Bridge Inspections.